ESTIMATING RADIAL MEASURES OF PRODUCTIVITY GROWTH - FRONTIER VS NON-FRONTIER APPROACHES

In this paper, we propose an alternative econometric framework for est imating and decomposing productivity change that does not require a di stribution for inefficiency or the uncorrelatedness between inefficien cy and the regressors, We develop our methodology for the input-orient ed radial measure of productivity change and establish that this equal s the negative of the time change in the log cost function. Our econom etric framework is based on a fixed-effects, multiple-output cost fron tier, where we decompose productivity change into discrete shifts in t he frontier and changes in firm efficiency levels relative to the fron tier. We also show that the standard non-frontier specification is nes ted within our frontier model and thus can produce different estimates of productivity change. Using a panel of twelve US railroads from 195 1 to 1975, our estimated cost frontier suggests average annual product ivity growth of roughly 0.3 percent, with efficiency change rising the n falling over the period. Specification tests reject the non-frontier model, which yields smaller gains in productivity.