The sensitivity of estimates of UK manufacturing TFP to definition and measurement
This thesis is concerned with the sensitivity of Total Factor Productivity (fFP) estimates in UK manufacturing industries to the use of alternative output concepts -gross output vs. value added- and estimation methods -growth accounting vs. econometrics. The departure point is that differences in methods and assumptions can lead to very different TFP growth estimates. The interpretation of these estimates can be problematic when they reflect factors beyond the theoretical concept of TFP. The central goal of this thesis is to evaluate whether and to what extent these factors have an impact on the measurement of TFP growth, on the estimation of any relationship between TFP and Research and Development (R&D) investment and, finally, on the measurement of the UK manufacturing productivity gap differential. The empirical results suggest that: First, TFP growth estimates in UK manufacturing are sensitive to both the output concept used and the assumptions underlying the method used to estimate them. It was found when tested for that the assumptions of perfect competition and instantaneous adjustment, which underlie the growth accounting framework, are not valid. Adjusting for the measurement bias associated with the presence of these factors, it is found that the recovery experienced in the 1980s in UK manufacturing productivity growth rates was not as spectacular as implied by the traditional growth accounting approach. Second, adjusting for measurement bias does not affect markedly the results found in related studies with respect to the relationship between TFP and R&D efforts. The results suggest that R&D investment from the industry itself and from other national industries has a positive impact on the industry's productivity but there is no gain from R&D investment undertaken abroad. Third, the results indicate that the bias in traditional TFP estimates does not impact greatly on the British productivity gap at the aggregate manufacturing level but does so at a more dis aggregated level. Finally, despite the concerns about measurement bias, the results show that the productivity gap still remains significant and the productivity of UK manufacturing still trails behind that achieved in the US, France and Germany, regardless of the sector.