THE DYNAMICS OF TOTAL FACTOR PRODUCTIVITY AND INSTITUTIONS

Abstract

(ProQuest: ... denotes formulae omitted.)1. INTRODUCTIONThe work of Solow (1957) and Abramovitz (1956) and more recent analyses (Casseli, 2005; Hall and Jones, 1999) demonstrate that total factor productivity (TFP) is the key driver of long-run income growth. Klenow and Rodriguez-Clare (1997) estimate that roughly 90 percent of the differences in of income per capita can be explained by differences in total factor productivity. It is also well documented that advanced economies (OECD) lead technological change and innovation while developing economies lag behind in the technological frontier and tend to adopt (with a lag) technologies developed in technology-leading countries (Besley and Case, 1993; Archibugiand and Pietrobelli, 2003). In addition, technologies created in leading countries may not be appropriate to be used in technology-backward economies (Basu and Weil, 1998; Acemoglu and Zilibotti, 2001). Thus, there are significant differences in levels and of productivity between advanced and developing economies.Studies examining cross-country TFP differences find strong evidence against global TFP convergence (Klenow and Rodriguez-Clare, 1997; Hall and Jones, 1999; Di Liberto et al. 2011). Di Liberto, Pigliaru and Chelucci (2011) show that most countries underperform respect to the U.S. in terms of TFP growth (p.168) as well as that the TFP gap across countries is persistent.1 While there is strong empirical evidence against global TFP convergence, there is evidence in favor of club convergence. Miller and Upadhyay (2002) group countries by income quartiles and find that there is absolute TFP convergence for countries in the lowest and highest income quartiles, but no convergence for countries in intermediate income quartiles. Kumar and Chen (2012) find that health and education have a significant positive effect on TFP and conditional TFP convergence. Papalia and Silvia's (2013) results also support club convergence. Madsen (2007, 2008) show that knowledge transmitted internationally through trade and patents has contributed significantly for TFP convergence among OECD countries. Di Liberto and Usai (2013) show that a polarization is taking place across European regions, with only a few regions emerging as TFP leaders while most regions are lagging behind, causing the TFP gap between these two clusters to widen.Loko and Diouf (2009) provide a comprehensive discussion of the factors that determine TFP and might explain the patterns (convergence, or lack thereof) discussed above. For the sake of simplicity, this study groups the factors affecting TFP into three categories. The first group consists of macroeconomic factors that either hinder or boost productivity growth. Economic instability (e.g inflation), a large government, and taxation distortions supposedly create market inefficiencies and, thus, negatively affect productivity (Barro, 1991; Loko and Diouf, 2009). On the other hand, overall openness to international trade and capital mobility are expected to boost productivity growth. International trade spurs competition - which leads to innovation - as well as serves as a channel for technology diffusion among nations. Thus, economies that are more open to trade are expected to have higher productivity (Dollar and Kraay, 2004; Wacziarg and Welch, 2008; Barro and Sala-i-Martin, 1995). The same rationale applies to capital flows. Openness to capital flows (Foreign Direct Investment) is associated with technology diffusion and knowledge transfers, which in turn boosts productivity (Borensztein et al. 1998). The composition of output (i.e, if intensive in services, agriculture, or manufacturing) has also been identified as a driver of productivity growth. In particular, nonagricultural economies have experienced faster productivity (Poirson, 2000; Jaumotte and Spatafora, 2007).The second group of factors includes variables that measure the quality of labor (human capital). …