Here’s a letter to Foreign Affairs. (HT to Luis Rubio for alerting me to Pettis’s piece.)
Editor:
Michael Pettis argues that, despite the failure of Smoot-Hawley tariffs nearly a century ago, Smoot-Hawley-like tariffs today will work (“How Tariffs Can Help America,” Dec. 27). The reason, Pettis insists, is that when the original Smoot-Hawley was enacted Americans consumed too little and exported too much, but today Americans consume too much and export too little.
Alas, this argument is a miasma of mistaken history and muddled economics.
It’s simply untrue, for example, that in the 1920s American income inequality rose, causing America to suffer, as Pettis writes, “from too much saving and too little consumption. It is why the country exported so much to the rest of the world.”
According to Nobel-laureate economist Simon Kuznets, the portion of total income paid as wages and salaries in the U.S. increased between 1919 and 1929 from 57.8% to 59.8%.* And over that same time, reports historian Burton Folsom, “the percentage of GNP that went to consumption expenses did not fall, but actually rose from 68 percent in 1920 to 75 percent in 1927, 1928, and 1929.”** Regarding exports: As an annual percentage of GDP from 1920 through 1929, merchandise exports (which back then included nearly all exports) averaged 5.5%. Today, exports as a percentage of GDP are double that rate, with an annual average from 2014 through 2023 of 11.7%. (The 2024 number is on track to be only slightly lower, about 11%.)***
As for his economics, Pettis naively describes tariffs as subsidizing production at the expense of consumption. Tariffs do indeed reduce consumption, but they don’t subsidize production. While tariffs might enable some firms to expand production, these duties reduce overall domestic production through at least three channels. One is the Lerner symmetry theorem (a basic economic concept apparently unknown to Pettis), which shows that a fall in imports is accompanied by a fall in exports. The second is the fact that today more than half of American imports are inputs used by American producers: Reduced access to these inputs reduces American production. The third channel is that, by diverting resources from tasks at which Americans have a comparative advantage to tasks at which Americans have a comparative disadvantage, American industry becomes less efficient and, hence, produces less. (And if foreign governments retaliate with their own tariffs, American production would fall even further.)
It’s distressing to find in your pages an essay on trade that would receive in an undergraduate economics course a grade of F.
Sincerely,
Donald J. Boudreaux
Professor of Economics
and
Martha and Nelson Getchell Chair for the Study of Free Market Capitalism at the Mercatus Center
George Mason University
Fairfax, VA 22030* Simon Kuznets, National Income and Its Composition, 1919–1938 (National Bureau of Economic Research, 1941), page 218.
** Burton Folsom, Jr., New Deal or Raw Deal? How FDR’s Economic Legacy Has Damaged America (Threshold Editions, 2008), page 35. See also Stanley Lebergott, Consumer Expenditures: New Measures & Old Motives (Princeton University Press, 1996).
*** See Our World In Data and FRED Data.
