Measuring the differences in productivities of Nations : a stochastic frontier approach

• Autor(a) principal: Aguiar, Diana Isabel Ribeiro
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.14/16993
• topic: Technical efficiency; Total factor productivity; Productivity growth; Stochastic frontier analysis; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

It is broadly accepted that differences in efficiency and productivity growth greatly contribute to the enormous differences in income across countries. Inefficiency levels were estimated for a panel of 40 countries, 34 of which are OECD-members and the remaining 6 are emergent economies, for the period of 2001-2011, using a stochastic frontier model based on the Battese and Coelli (1995) time-varying inefficiency model. Environmental variables were found to have an important role in explaining differences in technical efficiencies across countries. In particular, a high contribution of the agricultural sector and of natural resources rents to the economy, impediments to free trade such as tariffs, a bad business environment, a high number of patents, a high level of government debt and the financial crisis contribute negatively to technical efficiency. On the other hand, a good health status and good institutions help countries to be located closer to the frontier. Afterwards, productivity growth was decomposed using the Kumbhakar and Lovell (2000) primal frontier approach. The results showed that differences in TFP growth between developed and developing countries are the main drivers of the differences in the growth rates of GDP per worker, although differences in the factor accumulation also play an important role. Over the 2001-2011, we observed a general improvement in the technical efficiency of countries, which was outweighed by a downward shift in the stochastic production frontier.