GALTONIAN REGRESSION ACROSS COUNTRIES AND THE CONVERGENCE OF PRODUCTIVITY

Authors


  • I should like to thank Robert Barro, Simon Burke, John Cantwell, Mark Casson, Milton Friedman and Danny Quah for their comments on an earlier and longer version of this paper, Hart (1994), but they must not be held responsible for any of its shortcomings.

Abstract

ABSTRACTS

A Galton model, in logarithmic form is used to explain the convergence of countries' productivities over time, including β and σ convergence. The ‘regression fallacy’ does not arise. When comparing growths and levels of productivity, countries should be classified by their initial levels of productivity, preferably after logarithmic transformation. Errors in the data may justify using an error-in-variables model, or the geometric mean of the original and reverse regressions, but this is not certain.

Ancillary