My Account

Warning: You are using a browser that does not support angularJS. Some site functionality will not be available to you. Please consider updating to a newer version.

Paul Krugman Flunks Capital Theory

Sheldon Richman

Nobel laureate and New York Times columnist Paul Krugman is said to have bested commentator George Will over what prolonged the Great Depression during a joint appearance on ABC’s “This Week with George Stephanopoulos” back in November. But all Krugman really did was show that he, as a Keynesian, holds an unrealistic Play-Doh model of capital, as opposed to the more realistic heterogeneous, multistage, intertemporal structure-of-production model of the Austrian school of economics.

Here’s what actually happened. During the roundtable segment of the show, Will said, “[O]ne of the ways we turned a depression into the Great Depression . . .
was that there were no rules and investors went on strike because the government was completely improvising. Net investment was negative through almost all of the ’30s because, again, people did not know the environment in which they were operating because the government had the fidgets and would not let rules and markets work.”

Krugman responded, “Well, it’s not the way I read the history. . . . No, the negative net investment was because, you know, when you have 20 percent unemployment and all the factories are standing idle, who wants to build a new one? You don’t need to invoke the government to explain that.”

Point Krugman? Wrong.

If Krugman took the Mises-Hayek capital and trade-cycle theory seriously he’d realize that the idle factories in the 1930s represented malinvestment induced by Federal Reserve credit expansion in the 1920s. By lowering the interest rate and falsely signaling an increase in saving (that is, a preference for future over present goods), this policy shifted resources from later stages of production (closer to the consumer) to earlier stages of production. Unfortunately, those who think of capital as a heap of uniform, monochrome Play-Doh aren’t sensitive to this point. Capital is capital is capital. That’s why Krugman can’t understand why someone would want to invest in new facilities when others stand idle.

When the 1920s inflationary boom ended, as it had to because it was artificially induced and there weren’t enough resources for both the early stages and the later stages (where consumers wanted them), the malinvestments had to be liquidated and scarce resources had to be redeployed. But since capital consists not of malleable Play-Doh but rather of discrete things—buildings, machines, tools, materials—with particular characteristics, many of these products of malinvestment were unsuitable for other purposes. They couldn’t simply, costlessly, and instantly be moved and employed in later stages of production. Hence the idle factories. This was wasted capital brought about by the credit expansion. This was the Depression.

If the economy was to recover, new investment consistent with consumers’ actual preferences had to be undertaken. But that required time and saving—that is, deferred consumption, not the pumped-up consumer spending Krugman favors. It also required a stable political environment in which investors could be confident their property was safe from government predation. Unfortunately, thanks to tax increases, an unending stream of interventionist programs, and threatening antibusiness rhetoric, FDR’s government failed to provide that environment.

Krugman’s flip remark to Will is thus a perfect illustration of what is wrong with Keynesian economics. P.S. Will and Krugman believe it took World War II—“an enormous public works program,” in Krugman’s words—to end the depression. Both are wrong about that, as Robert Higgs documents in Depression, War, and Cold War. Ending unemployment with a military draft and boosting GNP through military contracts do not a recovery make. Living standards could hardly rise amid ration books, consumer-goods shortages, and war production.

Related Articles


{{}} - {{relArticle.pub_date | date : 'MMMM dd, yyyy'}}