Can Globalization Go Too Far?

Dani Rodrik, the political scientist’s favorite economist, argues for a limit to globalization in his recent book Globalization’s Paradox. The LSE EUROPP blog has a nice little summary of the book’s argument:

1. Markets require a wide range of non-market institutions (of regulation, stabilisation, and legitimation) in order to work well and remain socially sustainable.
2. These institutions do not take unique forms, in the sense that ultimate goals such as efficiency or stability can be achieved under a variety of designs and blueprints.
3. Different societies, organised around their own states, have patently different needs and preferences regarding the shape that market-supporting institutions can take.
4. A world that is sufficiently responsive to democratic preferences will therefore be one of institutional diversity and heterogeneity rather than institutional harmonisation and convergence.
5. Since institutional diversity inhibits the global integration of markets by raising transaction costs across jurisdictional boundaries, a world that is sufficiently responsive to democratic preferences will also be one that falls short of full globalisation.

The idea is that it is desirable to have different countries have different regulatory schemes, but full globalization demands complete harmonization of regulation in order to minimize transaction costs. Exporters and investors want to face the same regulations abroad as they do at home. Some of the premises of the argument are incorrect, but I actually agree with the conclusion for different reasons. First, the problems with the premises.

Premise #1 is disputable to some extent. It’s stated broadly enough that these “institutions” could be social or governmental, and the governmental institutions could be nothing more than enforcing appropriate rules of appropriation and exchange. Yet I suppose Rodrik means something more by the statement, as illustrated by his example of financial regulation. The sparer those institutions within which markets need to be embedded, the more harmonization is possible and desirable even if the rest of the argument is right.

Premise #3 is deeply problematic. It’s just not appropriate to speak of societies having “preferences,” and it’s even more naive to think of democracy as somehow translating those preferences into law. If the only reason for retaining national economic sovereignty is that the latter might permit different “societies” to enact their “preferences” into law, then the case is very weak.

My argument for an outer bound to globalization would be more contingent. Taking everything into account, the scales seem to tip against harmonization when: 1) different national approaches can give us better information about what works, rather than enshrining a subpar approach into law, 2) regulatory arbitrage allows better rules to develop, ultimately yielding harmonization but not through a centralized process, 3) smaller polities are less likely to be influenced by “insider” interest groups than some multilateral institution tasked with harmonizing regulations, or 4) citizens perceive globalization as a threat to sovereignty that they value, and therefore respond to harmonization attempts with a stronger backlash against globalization more broadly (Rodrik does address this last consideration in his book). Sometimes multilateral harmonization is justified and sometimes not; it depends on the balance of considerations. For instance, harmonizing customs rules, as was agreed recently by the WTO trade ministers in Bali, seems harmless.

Ultimately, I think Rodrik is pushing on a string. There is little risk that harmonization will go “too far,” given the extremely decentralized nature of global economic governance. And I disagree with Rodrik that globalization has gone about as far as it needs to go, and now requires defense from its own advocates. For instance, the West can do much more to repeal agricultural trade barriers that kill poor people around the world.