Friday, July 18, 2014

Evidence on the Samuelson Conjecture

"The Samuelson Conjecture" sounds a bit like the title of one of those old Robert Ludlum novels (The Bourne Identity, The Parsifal Mosaic, The Aquitaine Progression, The Sigma Protocol, etc.) But among economists, it refers to an article called "Where Ricardo and Mill Rebut and Confirm Arguments of Mainstream Economists Supporting Globalization," written by the great economist Paul Samuelson, and published in the Journal of Economic Perspectives in the Spring 2004 issue. (Full disclosure: I've been Managing Editor of the JEP since the first issue in 1987.)

Samuelson's article won't be a simple read for the uninitiated, but the basic intuition is straightforward.  The Samuelson conjecture starts from the basic insight that international trade is based on differences in productivity levels across industries that are driven by some mixture of natural endowments and past investments in education, physical capital, technology, and the legal and financial infrastructure. But what happens if a high-income and low-income economy start out in different places, but then the low-income economy develops in a way that it converges in education, investment, and technology toward the high-income economy? As the difference between the two economies decreases, the potential for gains from trade between the economies can also decrease. Or to bring this meditation abruptly to the real world, as the economy of China becomes more technologically similar to that of the United States, the standard of living in China undoubtedly rises, but the U.S. economy may suffer because the potential for gains from trade is diminished. Or as Samuelson puts it:
[S]ometimes free trade globalization can convert a technical change abroad into a benefit for both regions; but sometimes a productivity gain in one country can benefit that country alone, while permanently hurting the other country by reducing the gains from trade that are possible between the two countries. ... Historically, U.S. workers used to have kind of a de facto monopoly access to the superlative capitals and know-hows (scientific, engineering and managerial) of the United States. All of us Yankees, so to speak, were born with silver spoons in our mouths—and that importantly explained the historically high U.S. market-clearing real wage rates for (among others) janitors, house helpers, small business owners and so forth. However, after World War II, this U.S. know-how and capital began to spread faster away from the United States. That meant that in a real sense foreign educable masses—first in western Europe, then throughout the Pacific Rim—could and did genuinely provide the same kind of competitive pressures on U.S. lower middle class wage earnings that mass migration would have threatened to do. Post-2000 outsourcing is just what ought to have been predictable as far back as 1950. And in accordance with basic economic law, this will only grow in the future 2004–2050 period.
As Samuelson was at some pains to point out, this argument showing one possible theoretical outcome of how gains from trade can change as one country experiences technological development is not a new one; indeed, I often  have pointed out a version of this finding using the simple graphs in an introductory economics class. Samuelson was also at some pains to point out in his 2004 article that the argument didn't show that blocking imports from China was a good idea; after all, if the issue is how to keep gains from trade high, acting to reduce trade would be a peculiar way of accomplishing that goal. But for a few months after the article was published, a string of popular press articles made wildly overblown claims that the great Paul Samuelson had proven that globalization and international trade were harming the U.S. economy.

Theoretical models show what is possible, but empirical evidence shows the size of actual effects. In the American Economics Journal: Macroeconomics, Julian di Giovanni, Andrei A. Levchenko, and Jing Zhang have published "The Global Welfare Impact of China: Trade Integration and Technological Change" (6:3, 153–183). (Unlike the JEP, the AEJ-Macro is not freely available on-line, but many readers will have access through library subscriptions.) The authors do a simulation of the world economy, which is based on productivity estimates for many sectors across 75 countries: "We embed these productivity estimates within a quantitative multi-country, multi-sector model with a number of realistic features, such as multiple factors of production, an explicit nontraded sector, the full specification of input-output linkages between the sectors, and both inter- and intra-industry trade, among others."

The authors then focus on two scenarios for productivity growth in China. In one scenario, all of China's industries grow at the same productivity rate, so that the structure of China's economy remains fundamentally different from that of the U.S. In the other scenario, the productivity rate of China's industries grows in an unbalanced way, so that China's economy evolves in such a way that "China's productivity in every sector becomes a constant ratio of the world frontier."

Given the basic idea of international trade--that is, the economic gains from trade happen when trading across economies with different relative productivity levels across industries--one might expect that if productivity in China becomes more similar to U.S. patterns, then the U.S. economy could suffer from a reduction in the potential for gains from trade. But the simulation results from di Giovanni, Levchenko, and Zhang show the opposite results. For most countries, including the U.S., the gains from trade with China are larger when the patterns of China's productivity growth converges to the U.S. patter. The authors explain.
The sheer size of the Chinese economy and the breathtaking speed of its integration into global trade have led to concerns about the possible negative welfare effects of China’s integration and productivity growth. These concerns correspond to the theoretically possible, though not necessary, outcomes in fully articulated models of international trade, and thus have been taken seriously by economists. However, it is ultimately a quantitative question whether the negative welfare effects of China on its trading partners actually obtain in a calibrated model of the world economy with a realistic production structure, trade costs, and the inherently multilateral nature of international trade. ... With respect to technological change, our results are more surprising: contrary to a well-known [Samuelson] conjecture, the world will actually gain much more in welfare if China’s growth is unbalanced. This is because China’s current pattern of comparative advantage is common in the world, and thus unbalanced growth in China actually makes it more different than the average country.
Of course, one set of empirical simulations is not an unarguable proof, either, and research in this area is sure to continue. But for now, at least, the evidence suggests that Samuelson's (2004) conjecture emphasized a theoretical possibility that does not seem to hold true in a fully articulated quantitative model of world trade, in which the productivity gains in China, in the context of all the countries and trading relationships in the world economy, do redound to the overall benefit of the U.S. economy.

One final thought is that the discussion here focuses on the gains from trade as driven by differences in relative productivity  across countries, and these "Ricardian" models of trade have seen a new surge of importance in recent years. However, since the 1980s there have also been several waves of economic analysis that focused on other gains from international trade. For example, one approach focuses on gains to consumers from having a variety of products available (for example, a variety of models of cars or smartphone). A second approach looks at how international trade adds to the economic pressures that push less-productive firms out of business and provide benefits more productive firms that can sell in world markets. A third approach looks at how trade leads to pressure for innovation, which in turn increases productivity levels.

For a useful starting point to thinking about how the rise of emerging economies like China affect global trade, I recommend a symposium in the Spring 2012 issue of JEP. Gordon H. Hanson starts with "The Rise of Middle Kingdoms: Emerging Economies in Global Trade."  "Gains from Trade when Firms Matter," by Marc J. Melitz and Daniel Trefler, looks at the benefits of trade offers introduction to modern models of trade driven from variety, shifts toward more efficient firms, and technological gains. For an introduction to models of international trade based on by differences in relative productivity across countries--like the model used by di Giovanni, Levchenko, and Zhang--that same has a useful article called "Putting Ricardo to Work," by Jonathan Eaton and Samuel Kortum. Finally, Jonathan Haskel, Robert Z. Lawrence, Edward E.  Leamer, and Matthew J. Slaughter look at "Globalization and U.S. Wages: Modifying  Classic Theory to Explain Recent Facts."