Is population scale a worthless surrogate for business agglomeration economies?

Carlino argues that population scale has a negative effect on productivity because it is a diseconomy rather than an economy of agglomeration. This study arrives at different conclusions. After critiquing Carlino's arguments, empirical cross-sectional productivity equations show that capital intensity accounts for much of the urban productivity advantage. Measures of business services, localization effects, congestion, infrastructure, and labor quality are then included in the equations. These variables generally have the expected effect on productivity and are frequently significant. Nevertheless, population scale continues to have a positive effect in five industries; it develops a negative effect in only one.