What Paul Krugman Misses About 1937 Redux: Echoes

By Amity Shlaes 

What it if it just keeps going? That’s the question Americans are asking as they consider last month's 9.1 percent unemployment rate, still so high 33 months after the crash of September 2008. Scholars of economic history are asking another question: Are we repeating 1937?
That year, when Americans were expecting their economy to finally pull out of the Great Depression, the stock market dove again, with the Dow Jones Industrial Average dropping from the 190s in March 1937 to less than 100 in March 1938. Nonfarm private unemployment, the measure of Roosevelt's industrial economy, increased to more than 18 percent. Industrial production plunged by a third.
The problem then was monetary, some economists now say. Paul Krugman, in his New York Times column on June 2, argued that monetary and fiscal tightening caused the 1937 downturn, and might be squeezing the breath out of the economy now, precluding job creation. Krugman cites Gauti B. Eggertsson of the New York Federal Reserve Bank, who recently published blog posts and papers noting that the later 1930s, as now, saw higher commodity prices. Officials considered these rising prices a signal of inflation, and pressed for tightening. They erred.
This version of history holds, to some extent.
Or: Up to a point, Lord Krugman. After spending heavily during the 1936 election, President Franklin D. Roosevelt was now feeling more like a thrifty "Dutch householder," as the journalist Anne O'Hare McCormick wrote, and his administration cut executive-branch employees. His Treasury secretary, Henry Morgenthau, claimed that the government had "licked the Great Depression." Tax increases, themselves a form of fiscal tightening, were passed into law. In January 1937, Americans began to make their first payments into Social Security, taking away cash that workers might have spent on food, housing, clothing, liquor or cigarettes.
On the monetary side, the Fed doubled reserve requirements on banks, seeking to make them less vulnerable to failure. What the Federal Reserve didn't anticipate was that this would make the banks nervous, leading them to amass yet more reserves and taking money out of circulation. The Fed and the Treasury also fiddled with the gold-standard system for fear of inflation.
But two other factors are omitted from this narrative. The first is the price of labor. The Wagner Act, the great modern labor statute, became law in 1935. It made possible the closed shop, under which only unionized workers were allowed into a unit. In 1937, after Roosevelt was safely elected, labor leader John L. Lewis and his Congress of Industrial Organizations began using their new power to its full extent. Labor's tour de force in this period is memorialized in the photos we still recognize today of sit-down strikes at the General Motors Co. plant in Flint, Michigan. Strike days in 1937 totaled 28 million, up from 14 million during the election year.
Such labor stoppages, and the threat of more, led companies to raise wages more than they could afford to. Harold Cole of the University of Pennsylvania and Lee Ohanian of the University of California, Los Angeles, have demonstrated that wages in the latter half of the 1930s were well above trend for the entire century. Employers also hired less: Even as unionization increased, nonfarm unemployment did as well.
The second under-discussed issue is what scholar Robert Higgs has called "regime uncertainty." Roosevelt's victory in 1936 had been so convincing that people believed he might do anything. FDR reinforced this suspicion with an inaugural address so aggressive that modern presidential advisers would never allow it on the teleprompter. Roosevelt told the nation he sought in government "an instrument of unimagined power." That scared markets and small businesses.
Roosevelt relished hunting down big firms through regulatory action and blaming new sectors, such as utilities, for slowdowns -- on some days. Other days, he invited business leaders into the Oval Office and talked about partnership and a "breathing spell."
This inconsistency itself posed a problem. The diary of an Ohio lawyer named Daniel Roth, which was recently republished, captures the pervasive anxiety of the period. "We are having a bad steel strike in Youngstown and the mills have closed," Roth wrote on June 22, 1937. "The state and federal governments seem to support the labor unions and there has been a complete breakdown of law and order. Business is very quiet."
From the U.K., John Maynard Keynes wrote to FDR that it was all right to nationalize utilities or to leave them alone -- but what, Keynes asked, was "the object of chasing the utilities around the lot every other week?”"
In this respect, the 1938 midterm gains by Republicans were important because they signaled to Americans that there were limits to Roosevelt's power. And Roosevelt's decision to turn his attention to Adolf Hitler and Josef Stalin, and away from the supposed excesses of big business, contributed as much as anything else to the eventual recovery.
Addressing these last two areas in today's economy is not hard. To make employment less expensive, the government can undo the health-care law, our modern version of the Wagner Act, so that employers needn't worry that hiring implies accepting costs they can't control or even predict. Another boost to hiring would be a more reasonable National Labor Relations Board, not the current one, which chases companies such as Boeing Co. around the equivalent of Keynes's lot trying to drive up wages. Another would be to cut taxes for employers, big and small. Uncertainty would diminish if both parties publicly committed to a smaller and less intrusive government.
The problem isn't a single "Mistake of 1937" or "Mistake of 2011." It is the mistakes, multiple, of both periods.