Inequality and Technology: Did the Robots Do It?

Print
Monday, 21 January 2013 07:52

Adam Davidson has an interesting piece in the NYT Magazine on the debate over whether technology is responsible for the growth in inequality over the last three decades or whether the increase has been primarily the result of policies that have redistributed income upward. (As the author of The End of Loser Liberalism: Making Markets Progressive, I am firmly in the latter camp.) The immediate basis for the piece is a new paper by Larry Mishel, John Schmitt, and Heidi Shierholz that questions the widely accepted work of M.I.T. professor David Autor, which attributes rising inequality to the loss of jobs in middle class occupations. (The paper is not yet available, but several of the main points are presented in blog posts here, here, here, and here.)

Davidson does a good job laying out the central issues at one point turning to Frank Levy, another M.I.T. economist, to help define the terrain. Levy points out that while inequality has increased almost everywhere, there are huge differences in the extent of the increase. This suggests that there is a very big role for policy in the rise in inequality in the United States. (We're #1.)

However Davidson's conclusion may mislead readers.

"What do we value more: growth or fairness? That’s a value judgment. And for better or worse, it’s up to us."

The idea that there is tradeoff between growth and inequality does not follow from Levy's comments. It could be the case that policy decisions were aggravating trends in equality rather than alleviating them. For example, increasing the length and scope of patent and copyright protection is a policy that would have the effect of redistributing income upward as would protecting doctors and lawyers from international competition in a context where trade policy is designed to make most workers increasingly exposed to such competition. In these and other cases, it is possible to identify policies that would likely both increase growth and reduce inequality.