I hear many people, including some economists, say that a major cause of the U.S. Great Depression was the Smoot-Hawley tariff of 1930. I think their view of its cause is strongly affected by their justifiable hatred of trade barriers. In the past, when I’ve heard that, I’ve responded, “No, Smoot-Hawley was one of the causes, but it does not rank in the top three.”

The top three, in my view, are:

  • monetary policy, which Milton Friedman and Anna J. Schwartz argued was more important than any other cause;
  • Herbert Hoover’s and Franklin D. Roosevelt’s successful attempts to keep nominal wages in various sectors from falling; and
  • Hoover’s more than doubling of income tax rates.

I used to think that the Smoot-Hawley tariff was the fourth most important cause. But Douglas Irwin’s new book, Peddling Protectionism, has convinced me that Smoot-Hawley, though bad, was even less important than I had thought.

Why am I so convinced? Because Irwin, an economics professor at Dartmouth College and one of the world’s leading scholars of international trade, makes a careful, fact-freighted case. He points out that Smoot-Hawley did not raise tariffs to as high a level as is commonly thought. He also shows that international trade, so important to the U.S. economy today, was much less important then. And he shows that evidence presented by the late Jude Wanniski on the stock market’s reaction to Smoot-Hawley is quite weak. The one part of the commonly accepted view of Smoot-Hawley that holds up is the idea that it led to retaliation against U.S. exports by other countries’ governments.

Tariffs, considered | Consider tariff rates. Although the late Gottfried Haberler claimed in the 1970s that Smoot-Hawley had pushed tariffs to “skyscraper” heights, that is an exaggeration. It is hard to generalize about a bill that raised tariff rates on literally hundreds of separate items, but Irwin estimates that Smoot-Hawley raised the average tariff on imports subject to the tariff by 15–18 percent, an increase of about 6 percentage points. Irwin contrasts this increase with that caused by the Fordney-McCumber tariff bill of 1922, which raised the average tariff rate by a whopping 64 percent, or 13 percentage points. Of course, we had a boom in the 1920s, which is further evidence against the idea that tariff increases per se had much to do with economic decline.

We know that tariffs hurt an economy by reducing specialization and diverting domestic factors of production to high-cost items that we can get cheaper from other countries. Irwin does not argue that the tariff increases did not hurt the U.S. economy. But he claims that the damage was less than commonly thought — not only because the tariff increases were less than commonly thought, but also because there was not much trade on which to impose the tariffs. In 1929, imports were only 4.2 percent of gross domestic product. And that 4.2 percent substantially overstates the amount of imports subject to the tariff increases because, notes Irwin, only a third of imports, or 1.4 percent of GDP, were subject to tariffs to begin with. Smoot-Hawley raised tariff rates almost solely on goods already subject to tariffs. It is hard to believe that the 1.4 percent tail could wag the 98.6 percent rest of the dog.

Irwin notes that international trade did shrink after Smoot-Hawley. But if Smoot-Hawley was the main factor behind this shrinkage, reasons Irwin, then imports subject to the tariffs should have declined a lot, and duty-free imports should not have declined much. In fact, he finds, between the three months prior to Smoot-Hawley taking effect and the three months after, the value of “dutiable” imports fell by 34 percent, but the value of duty-free imports fell by a sizeable 21 percent. Using an equation for import demand, Irwin shows that only 4–5 percentage points of the 40 percent decline in imports was due to Smoot-Hawley, while a whopping 25–28 percentage points of the 40-point decline was due to lower U.S. income. What about the remaining 8–10 percentage points of the decline? Irwin attributes them to the fact that most tariffs were set in dollars and cents rather than as percentages of prices and, with massive deflation, tariff rates as a percent of import prices rose and would have risen even absent Smoot-Hawley. Irwin’s bottom-line estimate is that Smoot-Hawley reduced imports by about 0.2 percent of GDP, hardly enough to make a huge difference one way or the other.

Broader effects | The late Jude Wanniski, in his 1978 book The Way the World Works, argued, on the basis of stock-market declines, that Smoot-Hawley had a large negative effect on the U.S. economy. Wanniski wrote that the stock market declined by large percentages whenever the probability of passage of Smoot-Hawley increased. But Irwin carefully examines the evidence and finds Wanniski’s implausible. He notes that the Dow Jones Industrial Average fell by 6 percent on October 23, 1929, but that “the only tariff news that day was the failure of the coalition [in favor of Smoot-Hawley] to reduce the duty on carbide.” The Dow fell a whopping 23 percent on Monday and Tuesday, October 28 and 29, “but the only tariff news from Sunday was the statement by Sen. Reed that the tariff bill was dead and the response by Sen. Smoot that it was not.”

Irwin, moreover, quotes economist Scott Sumner’s research finding that while the stock market was plunging in October and November 1929, the widespread view at the time was that the protectionist wing of the Republican Party — the people pushing for Smoot-Hawley — was losing.

Irwin shows that critics of Smoot-Hawley are on firmer ground when they claim that other countries’ governments retaliated against U.S. exports. America’s largest trading partner then, as now, was Canada, and Canada’s government responded by reducing tariffs on British goods and raising tariffs on 16 products that, in total, accounted for 30 percent of the value of U.S. exports to Canada. He notes that after Canada’s government raised its tariff on eggs from 3 to 10 cents a dozen, American exports of eggs to Canada fell from 919,543 dozen to 13,662 dozen. After Spain’s government aimed a protectionist provision at American cars, U.S. car exports to Spain fell by 94 percent, while sales to Spaniards of British, Canadian, and German cars surged. Irwin argues that the worldwide depression would have caused many governments to increase trade barriers had Smoot-Hawley never been passed, but that Smoot-Hawley definitely contributed to the rise of protectionism, especially against U.S. goods.

And the bitter irony is that Smoot-Hawley did not work even on its own narrow terms and could not have worked. Why? Irwin points out that the bill was conceived by a farmer-dependent Republican Party as a measure to help farmers. But at the time, the United States was a large net exporter of cotton, wheat and other grains, and tobacco. In 1929, the United States exported $771 million of cotton and $286 million of wheat and other grains, but U.S. imports of those commodities were a measly $53 million and $20 million, respectively. And the areas of the agricultural economy in which the United States was a net importer were relatively small parts of the agricultural economy.

So, although Smoot-Hawley did not do as much damage as is often thought, it did do harm and did not accomplish its goal of helping farmers. Hmmm — causes damage and does not accomplish its primary goal. Does that sound like a lot of other government programs?