AMU - AMSE
5-9 Boulevard Maurice Bourdet, CS 50498
13205 Marseille Cedex 1
This work deals with the various methods of measuring total factor productivity (TFP) at the firm level (index numbers, data envelopment analysis, stochastic frontier, semi-parametric method and generalized method of moments). These methods are investigated both from a theoretical and empirical viewpoint with a panel of Turkish firms in order to highlight their similarities and dissimilarities. Our main results indicate that these methods do not estimate precisely the same components included in the TFP concept. Thus, the appropriate choice of the method appears to depend firstly on what we want to identify and secondly on the characteristics of data such as measurement errors and technology heterogeneity.
The aim of this work is to study the self-selection hypothesis on export markets assuming, firstly, that it is an exogenous process (random self-selection) and, secondly, that it is an endogenous process (conscious self-selection). Econometric estimates confirm recent empirical results according to which exporters are larger and more productive, and are involved in foreign markets. We also find strong evidence supporting the idea that self-selection is a conscious process. Thereby, firms increase their productivity with the aim of becoming exporters.
We test whether sunk exporting costs differ among markets. Our results confirm that during 1990-2002, sunk exporting costs were relevant for Spanish firms, and differed depending on the destination market. Besides, the costs to enter/re-enter were higher in developed markets.