I am a big fan of Dani Rodrik's writings on trade, and I agree with most of what he says in his NYT column today, but I do have one major disagreement. However, before going there let me emphasize some of the key points he makes in the piece.

First, Rodrik is very much on the mark in arguing that recent trade deals, like the Trans-Pacific Partnership, have very little to do with free trade. As he says, these deals are about imposing a corporate-friendly structure of regulations on both our trading partners and the U.S. (The deals have the effect of locking in laws that could otherwise be more easily altered.)

He also is right in singling out the pharmaceutical industry as the biggest villain in this story. We have been using these trade deals to ensure ever longer and stronger patents and related protections. The result is to make drugs, which would otherwise be cheap, extremely expensive. The price of drugs can be a serious burden even in rich countries, but patent protection can make life-saving drugs altogether unaffordable in developing countries. We should be looking to foster alternative, more efficient, mechanisms for financing research, not using trade deals to impose patent monopolies everywhere.

It's worth mentioning in this context the effort to impose rules on digital commerce in these trade deals. Folks following the scandals related to Facebook and Twitter's involvement in the presidential election know that we don't really have the rules down ourselves. In other words, we do not have a system in place that prevents both foreign and domestic actors from using dishonest means to influence public opinion and interfere with the democratic process. We also don't have effective systems in place to ensure the privacy of our personal data. These are really big issues that are probably worth getting sorted out before we try to shove a one-size-fits-all model on the rest of the world. 

Rodrick is also right to point out that the manufacturing jobs that have been lost due to NAFTA are not coming back pretty much regardless of what the outcome of the current negotiations. NAFTA might have been a bad deal for U.S. workers, but we can't run history backward. Once factories in the U.S. have been shut, it is very unlikely that companies will look to reopen them, even if the rules on NAFTA are changed. If the result of renegotiation is to make Mexico a less attractive place to manufacture items for the U.S. market, companies would more likely look to China or other developing countries as an alternative platform, rather than moving their operations back to the United States.

The place where I would disagree with Rodrik is his comment that:

"The overall trade deficit of the United States is determined primarily by macroeconomic factors — American consumers’ saving propensities, domestic corporations’ investment behavior and the federal government’s fiscal and monetary policies."

This is a view that sees U.S. output at being largely fixed and near its full employment level of output. In this story, a lower trade deficit would not mean increased output and employment, it would just mean that we would have less of the other components of demand (consumption, investment, or government spending). We could still care about the trade deficit since it may affect the mix of jobs, but it would not affect aggregate output.

This story might have been arguably true in the years prior to the Great Recession (not in my view, since I think we were often well below full employment levels of output), but it certainly has not been the case in most of the years since the Great Recession. (We are arguably close to full employment now.) Many mainstream economists, like Larry Summers and Ben Bernanke, accept the idea of "secular stagnation." This means that we faced a sustained shortfall in aggregate demand.

In that context, the trade deficit is not determined by the balance of national savings and national investment. In the world of secular stagnation, if we reduce the trade deficit, say by lowering the value of the dollar, this can lead to increased output and employment. The rise in output and employment leads to higher public and private savings. This means the standard national income accounting identities still hold (the trade deficit equals the gap between national investment and national savings), but the direction of causation goes from trade deficit to national savings, not the other way around.

This is an important point. People are not wrong to be worried about the trade deficit or to think it can be reduced through trade and currency policy (better the latter than the former). Our economy would have looked much better over the last decade if trade had been closer to balanced.



In an exchange Dani Rodrik indicated that he agreed about the impact of trade deficits. His piece was referring to the bilateral trade deficit with Mexico.