Tyler Cowen, whose site I recommend, has a fascinating piece up on an infamous moment in the history of the teaching of economics:
In the 1961 edition of his famous textbook of economic principles, Paul Samuelson wrote that GNP in the Soviet Union was about half that in the United States but the Soviet Union was growing faster. As a result, one could comfortably forecast that Soviet GNP would exceed that of the United States by as early as 1984 or perhaps by as late as 1997 and in any event Soviet GNP would greatly catch-up to U.S. GNP. A poor forecast--but it gets worse because in subsequent editions Samuelson presented the same analysis again and again except the overtaking time was always pushed further into the future so by 1980 the dates were 2002 to 2012. In subsequent editions, Samuelson provided no acknowledgment of his past failure to predict and little commentary beyond remarks about "bad weather" in the Soviet Union (see Levy and Peart for more details).Cowen goes on to suggest that there are reasons to discount the possibility that ideology influenced Samuelson in making this mistake. I'm not sure I buy his arguments there and have commented to that effect on his site. What interests me here is the larger question of the legitimacy of Economics as a prediction-making science.
The Samuelson textbook is not any textbook, it dominated the market like no other. And yet here is this glaring predictive mistake and not only Samuelson but seemingly no one else was much bothered by it at the time. Samuelson didn't just miss the boat here, he wasn't even in the right ocean. This isn't an indictment of Samuelson, it's an indictment of the entire field of study.
And it feeds back into one of Alsadair MacIntyre's more powerful arguments: Whatever else social scientists such as economists have or have not done, they have not demonstrated an ability to make useful predictions that is even close to what we expect of other sciences.
No comments:
Post a Comment