Should Slowing Trade Growth Worry Us?
I’ve spent most of today both under pressure to get an assignment out the door and under the weather; still sniffling, but the piece has been emailed off, so a bit of time for the blog. Except I feel like taking a vacation from both the shutdown and Obamacare. So let’s talk about trade — specifically, a recent post by Gavyn Davies, “Why world trade growth has lost its mojo,” which expresses deep concern over the fact that in recent years trade hasn’t grown much faster than global GDP. He suggests that hidden protectionism may be partly to blame, and that this may have large economic costs.
So, I’m going to disagree with both propositions.
First, on the general point of the welfare gains from trade: I’m basically with Dani Rodrik here. Standard economic models do not imply huge gains from trade liberalization. You can make arguments that suggest bigger gains, but they’re highly speculative, and the credulity with which people accept dubious nonstandard arguments for big trade gains contrasts oddly with the gimlet eye cast on arguments for, say, industrial policy. You should definitely not accept estimates that every dollar of additional trade raises world GDP by 46 cents — an extremely high number — as being definitive.
But my main thought, reading Davies’s piece, was that the belief that trade must always expand much faster than output, and that there’s something wrong if it doesn’t, doesn’t stand up to careful scrutiny.
In part, this notion comes from the fact that trade has grown faster than output since 1950. However, up through about 1970 that only represented a return to levels of trade relative to output that prevailed before World War I:
On the other hand, one does see that business cycle fluctuations produce large fluctuations in trade, much bigger in percentage terms than the moves in GDP, which you might take — which Davies does take — as an indication that the “income elasticity” of trade, the percentage rise for every percent rise in GDP, is much bigger than one.
This is, I’d say, a confusion between short-term and long-term issues. Consider, instead of trade, industrial production. We know that this fluctuates much more than GDP over the business cycle, because purchases of manufactured goods slump much more in recessions than purchases of services. Over the long run, however, industrial production and GDP grow at roughly equal rates. There’s no reason trade couldn’t be the same way. In fact, one reason trade fluctuates so much in the short run is precisely because it’s dominated by manufactured goods.
To explain a rising long-term ratio of trade to GDP, we have to turn instead to structural changes in the world economy, of which the most obvious involve declining costs of trade. My view is that rapid trade growth since World War II was driven by two great waves of trade liberalization and one major technological innovation. The first wave of trade liberalization involved industrial countries, and was largely over by 1980:
The second wave involved the great opening of developing countries:
World Bank
This is still going on, but the major opening of Latin America, China, and India is already well behind us.
Finally, there’s The Box — containerization, which made the vertical disintegration of production, with separate stages carried out in far-distant nations, possible. But this too has been going on for a while.
The point is that it’s entirely reasonable to believe that the big factors driving globalization were one-time changes that are receding in the rear-view mirror, so that we should expect the share of trade in GDP to plateau — and that this doesn’t represent any kind of problem. In fact, it’s conceivable that things like rising fuel costs and automation (which makes labor costs less central) will lead to some “reshoring” of manufacturing to advanced countries, and a corresponding decline in the trade share.
Ever-growing trade relative to GDP isn’t a natural law, it’s just something that happened to result from the policies and technologies of the past few generations. We should be neither amazed nor disturbed if it stops happening.
So, I’m going to disagree with both propositions.
First, on the general point of the welfare gains from trade: I’m basically with Dani Rodrik here. Standard economic models do not imply huge gains from trade liberalization. You can make arguments that suggest bigger gains, but they’re highly speculative, and the credulity with which people accept dubious nonstandard arguments for big trade gains contrasts oddly with the gimlet eye cast on arguments for, say, industrial policy. You should definitely not accept estimates that every dollar of additional trade raises world GDP by 46 cents — an extremely high number — as being definitive.
But my main thought, reading Davies’s piece, was that the belief that trade must always expand much faster than output, and that there’s something wrong if it doesn’t, doesn’t stand up to careful scrutiny.
In part, this notion comes from the fact that trade has grown faster than output since 1950. However, up through about 1970 that only represented a return to levels of trade relative to output that prevailed before World War I:
This is, I’d say, a confusion between short-term and long-term issues. Consider, instead of trade, industrial production. We know that this fluctuates much more than GDP over the business cycle, because purchases of manufactured goods slump much more in recessions than purchases of services. Over the long run, however, industrial production and GDP grow at roughly equal rates. There’s no reason trade couldn’t be the same way. In fact, one reason trade fluctuates so much in the short run is precisely because it’s dominated by manufactured goods.
To explain a rising long-term ratio of trade to GDP, we have to turn instead to structural changes in the world economy, of which the most obvious involve declining costs of trade. My view is that rapid trade growth since World War II was driven by two great waves of trade liberalization and one major technological innovation. The first wave of trade liberalization involved industrial countries, and was largely over by 1980:
World BankFinally, there’s The Box — containerization, which made the vertical disintegration of production, with separate stages carried out in far-distant nations, possible. But this too has been going on for a while.
The point is that it’s entirely reasonable to believe that the big factors driving globalization were one-time changes that are receding in the rear-view mirror, so that we should expect the share of trade in GDP to plateau — and that this doesn’t represent any kind of problem. In fact, it’s conceivable that things like rising fuel costs and automation (which makes labor costs less central) will lead to some “reshoring” of manufacturing to advanced countries, and a corresponding decline in the trade share.
Ever-growing trade relative to GDP isn’t a natural law, it’s just something that happened to result from the policies and technologies of the past few generations. We should be neither amazed nor disturbed if it stops happening.
