Bailout

Dismal science datapoint of the day

By Felix Salmon

Alex Tabarrok blogs a recent paper by David Levy And Sandra Peart:

Like Samuelson, McConnell estimated Soviet GNP as half that of the United States in 1963 but he showed that the Soviets were investing a much larger share of GNP and thus growing at rates “two to three times” higher than the U.S. Indeed, through at least ten (!) editions, the Soviets continued to grow faster than the U.S. and yet in McConnell’s 1990 edition Soviet GNP was still half that of the United States!

As Levy and Peart note, this isn’t a function of leftist ideology: more left-wing textbooks failed to make the same mistake. Rather, it’s just a trust in models. Tabarrok explains:

When the war ended, the fundamental tradeoff became one between investment and consumption and since the Soviets invested a greater share of GNP they would naturally consume less but grow faster. Moreover, since the Soviets had solved the unemployment problem they were, if anything, more efficient than the U.S.

Levy and Peart conclude that economics needs more competition among models. I’d conclude that economics textbooks need that too — along with a healthy skepticism as to the real-world utility of any given model.

Full post as published by Macroprudential on January 04, 2010 (boomark / email).