Presented by: 
Chris O'Donnell, UQ School of Economics
Tue 13 Aug, 11:00 am - 11:45 am

This paper shows how a short time-series can be used to compute the Fare-Primont total factor productivity index recently proposed by O'Donnell (J Prod Anal, 2012b).  The Fare-Primont index seems to be the only total factor productivity index that can be constructed from distance functions without violating important transitivity and identity axioms from index theory.  It also has the important property that it can be exhaustively decomposed into various measures of environmental change and efficiency change.   This paper uses Bayesian methodology to effect this decomposition.  Bayesian methodology is used partly because  it can solve an endogeneity problem that arises in the econometric estimation of distance functions, and also because it can be used to make valid small-sample inferences concerning measures of efficiency.  Results for the Australian Northern Prawn Fishery for the period 1974--2010 are summarised in terms of characteristics of estimated posterior pdfs.  Variations in productivity and scale efficiency change are found to be strongly linked to changes in fishery management.

Visit Prof. O'Donnell's webpage (new window)

Return to main SMOR seminar                   webpage