In a recent interview Katherine Tai says: “Our industrial policy is one that has to work with trade policy.”
I would put it even more strongly. I would say that there is no fundamental difference between industrial policy and most forms of trade policy. An industrial policy must have trade effects because all industrial policies affect the relationship between savings and investment.
It’s worth noting that the most mercantilists economies in the world, with the biggest trade surpluses, rarely if ever see themselves as protectionist or as having implemented trade policies. They usually think of their polices as industrial polices designed mainly to raise manufacturing employment by putting into place policies that increase the competitiveness of the manufacturing sector. Their large, persistent surpluses, they claim, are simply the consequence of efficient manufacturing and household thrift.
That’s nonsense, of course. Trade policies don’t need to involve tariffs. They may instead involve administered lending, heavy spending on logistics and transportation, “stable” currencies, weak labor protection, minimal social safety nets, tax rebates, R&D spending, etc. As with tariffs and other explicitly mercantilist policies: these policies change the distribution of income in ways that can affect the relationship between savings and investment or, which is the same thing, the relationship between domestic production and domestic demand.
To the extent that countries implement trade or industrial policies that reduce domestic demand and externalize the costs through trade surpluses, other countries must also implement trade or industrial policies if they want to protect themselves from absorbing the cost. That’s why the US and other countries must implement policies – whether they call them trade policies or industrial policies – that either reverse their deficits or that encourage government entities to turn their deficits into fund productivity-enhancing investments.
That’s also why I have little faith in WTO-style agreements that target a specific set of policies as “mercantilist”. There are an infinite number of such policies and so they cannot all be constrained. It would be impossible to do so. What matter ultimately is whether or not in the aggregate these policies result in large, persistent trade surpluses. If they do, they are effectively beggar-thy-neighbor policies that increase domestic growth at the expense of trade partners. That is why trade agreements should target trade imbalances, and not the specific policies that might or might not lead to trade imbalances. We need a systemic approach to trade. Legalistic or incremental approaches make little sense.
svembu comment
Prof Pettis makes the extremely important point that persistent trade surpluses by themselves are evidence of mercantalism/ protectionism that WTO does not grasp. Countries that run persistent trade deficits must guard against them. Great thread.
Chinese policy can be summarized as “compulsive overproduction”. This is not good for China, and it is not good for the world. In India, it has caused import dependence, even “addiction”. This the new Opium War, this time waged by China. This is not a free market phenomenon.