This paper, in contrast to much of the existing literature, which focuses on the impact of Information and Communication Technologies (ICT) on labour productivity, assesses the relationship between ICTs investments and Technical Efficiency (TE) using a stochastic frontier approach. We use a large panel dataset of Italian manufacturing firms over the period [1995][1996][1997][1998][1999][2000][2001][2002][2003][2004][2005][2006] and confirm prior findings of the existing literature on ICT and productivity. In addition, we test on which extend the ICT investments influence the distance of the firm from the production frontier; that is, how ICT's adoption influences the closing of the firm efficiency gap. We also test how long the effects of ICT investments on technical efficiency last. We find that ICT returns on TE are influenced by some firm's characteristics, most of them idiosyncratic, such as management practices, labour organization, research and development.Keywords: ICT; stochastic frontier; technical efficiency; manufacturing firms JEL: D24, L25, L63, O33 1
IntroductionAfter a long debate on the returns of investment in information and communication technology (ICT) there is now general agreement that ICT positively contributes to economic growth both at micro and macroeconomic levels. This debate has been largely based on the productivity paradox, which started soon after Robert Solow's (1987) statement: "You can see the computer age everywhere but in the productivity statistics".The related empirical literature studies both the relationship between ICT investments and labour productivity and ICT investments and total factor productivity (TFP). Few attempts have been made to also study the relationship between ICT investments and technical efficiency (TE) at firm level. The importance of this relationship arises from the fact that the productivity growth is mainly the result of technical and efficiency changes. Hence, it is important to verify the effect that ICTs have both on productivity and TE.In this paper the Cobb-Douglas and Translog production functions are used to explore ICT investments impact on firm distance from the 'best practice technique'. We utilize the stochastic frontier model introduced by Battese and Coelli (1995). This methodology has the benefit that it uses a one-stage procedure to estimates both productivity and (in)efficiency. The paper adds to the existing literature in four ways. Firstly, ICT capital and high-skilled workers are considered as inputs in the firm's production function. Secondly, ICT investments are considered as a factor able to directly influence TE. Thirdly, we investigate the length of the positive impact of ICT on firm efficiency. Finally, we postulate that ICT effects on firm efficiency depend on some complementary idiosyncratic factors (such as management practices, research and development investments and other firm' characteristics) that are able to boost ICT returns.The analysis is conducted using a balanced and an unbalanced panel data of Itali...