The Gravity ModelThe gravity equation is the most empirically successful equation in economics (Anderson & Wincoop, 2001). It has been used regularly to estimate the impact of exchange rate volatility, border effects and the impact of trade agreements between countries and regions. The gravity model relates to bilateral trade flows to GDP, distance and other factors that affect trade barriers. It is used to deduce trade flow effects of exchange rates, custom unions and global boundaries. Economists have found out that after controlling for size, trade between two regions decreases in their bilateral trade barrier in comparison to the average barrier of the two regions to trade with all their partners.
A region that is resistant to trade with all others is forced to trade with a given bilateral partner. Economic analysts call this theoretically appropriate average barrier the multilateral resistance. The gravity model was originally under disregard because it had no theoretical basics. There are two gravity models. One such model is the theoretically derived gravity model while the other is an empirically based gravity model. Empirical gravity models do not have theoretical basics.
Empirical gravity literature does not include any form of multilateral resistance in analysis. It includes non theoretical distance variables that are related to distance, to all bilateral associates. The model’s remoteness index does not take into account any of the other trade obstacles that are the focus of analysis. This paper provides a critical examination of the superiority of the theoretically derived gravity model over the empirically based gravity model. It also explains the inferences of the theoretically derived gravity model on applied econometric analysis of trade flows.
Superiority of the Theoretically Derived Gravity Model over the Empirically Based Gravity ModelFirstly, the theoretical gravity model is equipped with theoretical foundations. These theoretical foundations equip the model with a strong explanatory power. This has been the persuasive motivation for the model’s usage. Trade economists have written papers that take theory seriously, but the papers are usually viewed as contributions to narrow empirical topics such as the size of the border effect, or the level of the elasticity of a substation. Therefore, the methodological advances in these papers have been ignored in literature.
Comparative statistics for applied econometric analysis of trade flows cannot be done by use of equations alone. There should be an excellent theory that explains the equations (Baldwin & Taglioni, 2006). This has led to the formulation of biased quantitative, long run effects of Free Trade Agreements on trade flows, using the standard cross section gravity equation. The empirically based gravity model ignores unobservable heterogeneity, and this leads to biases in estimates. This bias, has led to a significant underestimation of conventional estimates of the effect of Free Trade Agreements on bilateral trade flows.
A theoretically derived gravity model provides the most reasonable estimates of the average effect of Free Trade Agreements on a bilateral trade flow. This is because estimates are obtained from a theoretically motivated gravity equation using panel data with bilateral fixed, and country and time effects or differentiated panel data with country and time effects. The empirically based gravity model fails to identify the impact, because it estimates variables using cross sectional data. This is compromised by a lack of suitable instruments (Baier & Bergstrand, 2005).