Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
A number of analytical contributions have given sufficient conditions under which Leontief multiplier estimates are biased. This paper uses empirical data and a Monte Carlo framework to evaluate the problem of bias and obtains the opposite conclusions. For practical purposes, it appears that linear multiplier estimates are unbiased. The variance properties of these estimates are also evaluated. Copyright 1989 by MIT Press.