A commentary on the gravity equation in international business research

Abstract

PurposeThis commentary aims to discuss methodological issues and practical concerns of international business scholars who apply the gravity equation in their research.Design/methodology/approachThe paper summarizes and compares theories and several advanced empirical specifications in the gravity equation.FindingsThe paper proposes a relatively low‐cost specification and estimation to implement such correction, which is robust in the presence of various endogeneity effects and non‐stationary variables. In the presence of zero‐values in the dataset, however, the multilateral specification is best estimated with a Poisson maximum likelihood.Originality/valueThe most important message of the commentary is that this equation should correct for multilateral resistance factors.

Trade Costs and Economic Development

“…The alternative to the second downside was provided by Carrère (2006), who modeled pair dummy variables as random variables (RE). By using pair REs, we could still apply Baldwin and Taglioni's (2006) approach to handle the multilateral resistance factor, but, in addition, we could estimate the impact of distance on trade (Fratianni, Marchionne, and Oh 2011).…”

“…The correlation coefficients shown in the matrix are high, suggesting that our analysis is robust to different methodologies. We also used other estimators, such as the negative binomial, pseudo‐Poisson maximum likelihood estimator, and feasible generalized least squares, with similar results; for space reasons, the results are not shown (for a comparison of different estimators in the GE, see Fratianni, Marchionne, and Oh 2011).…”

Impactos da facilitação de comércio: evidências do modelo gravitacional

“…VAN WiNCOOP, 2003;FEENSTRA, 2004;FRATiANNi, 2007 (2007), o presente estudo estima o modelo gravitacional considerando-se os efeitos fixos bilaterais. A utilização de efeitos fixos nos pares de países capta, além dos termos de resistência multilateral, os custos de comércio bilaterais não observados diretamente, tais como os custos de transporte, particularidades geográficas e fatores históricos (CHENG e WALL, 2005).…”

Is Our World Going to Get a Whole Lot Smaller?

“…11 Our estimation approach takes into account two recent methodological innovations in the gravity model literature. Egger (2002), van Wincoop (2003, 2004) and Anderson (2011) have shown that omitting controls for multilateral resistance can create severe biases on distance estimates in a gravity setting.…”