Has Distance Died? Evidence from a Panel Gravity Model

Abstract

This paper reports panel gravity estimates of aggregate bilateral trade for 130 countries over the period 1962-96 in which the coe±cient of distance is allowed to change over time. In a standard speci¯cation in which transport costs are proxied by distance only, it is found paradoxically that the absolute value of the elasticity of bilateral trade to distance has been signi¯cantly increasing. The result is attributed to a relatively larger decline in costs independent of distance (such as handling) than in distance-related costs (e.g. oil price). An extended version of the model that controls for these two factors eliminates this positive trend without reversing it. However, when the sample is split into two groups (`rich-rich' and`poor-poor'), the paradox is maintained for the`poor-poor' group. While not conclusive, these results are consistent with the view that poor countries may have been marginalized by the current wave of globalization.

Distance, globalization, and international trade

“…Following the recommendations of Brun et al (2005) and Piermartini and Yotov (2016) we will estimate equation (2) with panel data, using exporter-time and importer-time fixed effects in order to control for any possible characteristics on the exporter and importer side that may influence bilateral trade. 8 The panel approach will enable us to trace the evolution of estimated effects of distance and any other gravity variables of interest within the same econometric specification.…”

“…However, it is likely that these forces have not penetrated countries uniformly, cf. Brun et al (2005) and Carrère et al (2013). Our goal in this section is to test this argument by allowing for heterogeneous distance effects across countries in our sample with a view to establishing which (groups of) countries benefitted the most from globalization.…”

“…See Buch et al (2004), Carrère and Schiff (2005), Brun et al (2005), Boulhol and de Serres (2010), Lin and Sim (2012), Yotov (2012) Carrère et al (2013) and .…”

International Migration: A Global Complex Network

“…As the main aim of this paper is to estimate the impact of the different variables on migration flows, it is preferred to use an RE model rather than an FE model, as because of the inherent first differentiation process, the latter will result in the elimination of the time invariant explanatory variables which are vital in our analysis (e.g. Brun et al, 2005;Etzo, 2011). …”

Which commercial partners are important for the most recently admitted EU countries?

“…In some studies, missing values are treated as zeroes (Gleditsch, 2002;Brun et al, 2005;Felbermayr and Kohler, 2006;and Coe et al, 2007), while others insists that these values are conceptually di¤erent (Harris et al, 2012). In Sections 4 and 5 we addressed the zero values problem by adding one unit to the trade ‡ow data before taking logarithms (Chen, 2004), while missing values were excluded from the sample.…”