History, regionalism, and CARICOM trade: A gravity model analysis

Abstract

A gravity model is fitted to 17 annual cross sections, 1980-1996, of bilateral trade data (imports and exports) between CARICOM member states and selected trading partners. Specifically, the paper investigates the effects of regional integration, colonial legacies and linguistic ties on CARICOM bilateral trade flows between member states and between member states and North American and European Union countries. The empirical results indicate that history and regionalism have had significant and strong effects on CARICOM trade. Additionally, two traditional explanatory variables in gravity models, income per capita and population, have significant and positive effects on CARICOM trade while the distance between trading partners exerts the expected negative effect.

The Effects of Trade Facilitation on Trade Performance in Africa

“…Turning to the control variables in the gravity model, the coefficients are within plausible range of standard gravity model estimates. The trade elasticity of GDP indicates a positive and significant GDP effect on trade for a both exporting and importing countries comparable to those reported by Sandberg et al (2006). Similarly, the population variables are positive, though not significant at conventional levels of statistical significance.…”

“…Traditionally, the gravity model is specified in a naive form where the GDP and distance variables are augmented with observable trade cost variables such as trade agreements, common currency, contiguity and common language (see Sandberg et al, 2006). However, following the works of Anderson and van Wincoop (2003), the emphasis has been placed on the Multilateral Resistance Term (MRT).…”

Trade facilitation and its effects on Chile’s bilateral trade between 2006 and 2014

“…Recently, there has been an increase in the number of articles in which the gravity model is applied to determine export behaviour, which, directly and significantly, include Chile or the Southern Cone countries, the Southern Common Market (MERCOSUR) or the Andean Community (CAN) (Durso and Ochoa, 2003;Giacalone, 2003;Kamil and Ons, 2003;Martínez-Zarzoso and Nowak-Lehmann, 2003;Nicita, Olarreaga and Soloaga, 2003;Vallejo and Aguilar, 2004;Cárdenas and García, 2004;Lewer and Sáenz, 2004;Martínez-Zarzoso and Suárez-Burguet, 2004;Lara and Soloaga, 2005;López and Fernando, 2005;Sandberg, Seale and Taylor, 2006;Sá Porto and Azzoni, 2007;Serrano and Pinilla, 2008;Valenzuela-Klagges, 2011;Ramos-Martínez and others, 2012;Álvarez, Fischer and Natera, 2013;Florensa and others, 2013;Bacaria-Colom, Osorio-Caballero and Artal-Tur, 2013;Hernández, 2014;Valenzuela-Klagges and Espinoza-Brito, 2015). Several of these studies use cross -section data to apply this model; but nowadays there is a tendency to estimate the gravity model using static and panel data with fixed or random effects Burnquist, 2011, andSant' Anna andDe Souza, 2014) and dynamic effects.…”

La facilitación comercial y sus efectos en el comercio bilateral chileno de 2006 a 2014

“…En los últimos tiempos se ha incrementado la cantidad de artículos en los que se aplica el modelo gravitacional para determinar el comportamiento exportador, y en los que se incluye, de forma directa y significativa, a Chile o a países del Cono Sur, el Mercado Común del Sur (MERCOSUR) o la Comunidad Andina (CAN) (Durso y Ochoa, 2003;Giacalone, 2003;Kamil y Ons, 2003;Martínez-Zarzoso y Nowak-Lehmann, 2003;Nicita, Olarreaga y Soloaga, 2003;Vallejo y Aguilar, 2004;Cárdenas y García, 2004;Lewer y Sáenz, 2004;Martínez-Zarzoso y Suárez-Burguet, 2004;Lara y Soloaga, 2005;López y Fernando, 2005;Sandberg, Seale y Taylor, 2006;Sá Porto y Azzoni, 2007;Serrano y Pinilla, 2008;Valenzuela-Klagges, 2011;Ramos-Martínez y otros, 2012;Álvarez, Fischer y Natera, 2013;Florensa y otros, 2013;Bacaria-Colom, Osorio-Caballero y Artal-Tur, 2013;Hernández, 2014;Valenzuela-Klagges y Espinoza-Brito, 2015). En varios de estos estudios se utiliza el corte transversal para aplicar este modelo, pero en la actualidad se observa una tendencia a estimar el modelo gravitacional mediante datos de panel estáticos con efectos fijos o aleatorios (De Souza y Burnquist, 2011, y Sant'Anna y De Souza, 2014), y dinámicos.…”