Empirically Based Gravity Model - Example

INTERNATIONAL MARKETS, INSTITUTIONS AND POLICY] (Name) (Instructor/Tutor) (Course/Subject) (Institution/ University) (City, State) (Date) The Gravity Model The 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 Model Firstly, 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). Secondly, the theoretically derived gravity model explains most of the variation in international trade (Rose, 2004). It has a natural ability to explain a large fraction of variations in an observed volume of bilateral trade. This is because it can be justified by a variety of theories such as Heckscher-Ohlin model with specialization and the monopolistic completion theory. The estimation results that this model generates are not biased because of omitted variables, as is the case with the empirically based gravity model (Subramanian & Wei, 2006). Also, economists can conduct a comparative statistics using the theoretically derived gravity model. In fact, conducting comparative statistics is the purpose of estimating gravity equations. The theoretically derived gravity model is superior because the estimated effects of distance and output that it generates are sensible and economically and statistically significant. A theoretical gravity model can be used to analyse the effect of regional trade blocs and currency unions. Unlike the empirically based gravity model, it does not ignore a multilateral resistance or remoteness term. Anderson and Wincoop (2001), suggest that the inclusion of country fixed effects by the theoretically derived gravity model captures multilateral resistance, and corrects misspecifications that could have been created by the empirically based gravity model. The gravity model requires fixed effects for both importers and exporters. This is because trade between any two countries depends on the multilateral resistance of both importers and exporters. Empirically based gravity models do not include country fixed effects. Therefore, the empirically based gravity model lacks a logical conclusion because it does not include time-varying importer and exporter fixed effects. The theory derived gravity model provides predictions that a properly specified gravity model finds consistent evidence. The theoretical gravity model allows economists to make inferences about unobservable trade costs. This is because links trade costs to observable cost substitutes and makes an assumption about error terms that link observable trade flows to theoretically predicted values (Anderson & Wincoop, 2004). Observable arguments that have been used in the theoretically derived gravity model include directly measured trade costs, distance, and adjacency, preferential trade membership and common language. This model can be used to analyse the impacts of regional trade agreements and the effect of presumptively implausible regularity restrictions on nontariff barriers and customs union membership upon trade volume. Each idea can be applied to aggregation, in several dimensions to provide answers to various questions. When using gravity models, it is accepted to reflect on ideal aggregation over trading partners such as aggregation over commodities. However, the theoretically derived gravity model has some pitfalls in the creation of assumptions. For instance, the model presumes that all countries purchase goods from all suppliers. This might lead to formation of incorrect or biased inferences. The model can account for zero bilateral trade flows at a disaggregated level, involving small regions or developing countries. The theoretically derived gravity model takes into account, the differences in regions and preferences .The empirically based gravity model assumes that preferences and technology are the same for all agents. Differences in preferences are, however, empirically indistinguishable from trade costs. The theoretically derived gravity model is superior because it can be used with trade cost that cannot be measured directly. A theoretical approach is inevitable to surmise the large portion of trade costs that cannot be directly measured in the data. The literature on deduction about trade barriers from final commodities prices is a devoid of theory, and requires effective use of trade theory to fill the gap. The theoretically derived gravity model uses evidence on prices. This has filled the gap between practice and theory in the inference of trade costs from trade flows. Therefore, the theoretically derived gravity model provides the main connection between trade barriers and trade flows. The empirically based gravity model is considered atheoretic while theoretically derived gravity model is justifiable under restrictive assumptions. The theoretically derived gravity model has been extremely successful empirically, and useful as the basis for tests of other propositions (McCallum, 1995). The theoretically derived gravity model is superior because it forms the basis of empirical analysis. It provides a theory that economists can use in analysis of the effects of bilateral trade elements. Empirical analysis cannot be done without using a clear theoretical framework. There should be a link between theory and data. Theory provides an excellent story that supports the high partial correlations that empirically based models make (Leamer & Levinsohn, 1994). Implications of the Superiority of the Theoretically Derived Gravity Model on Applied Econometric Analysis of Trade Flows Applied econometric analysis of trade flows involves conducting of comparative statics. This includes finding out the effects of removing certain trade barriers on bilateral trade between countries or regions. Economists ask questions regarding effects of the removal of trade barriers such as the creation of Free Trade Agreements and constructing of trading blocs. Economists should be able to solve the general balance model before and after the removal of trade obstacles. The theoretically derived gravity model has various impacts on applied econometric analysis of trade flows. First, a theoretically grounded approach can be used validly to compute the impact of borders on trade. This involves trade between a country’s regions and international partners. This is possible because there is always an underlying theory that explains the assumptions regarding trade volumes and characteristics of various regions or states. Empirically based gravity models cannot be used to compute such impacts reliably because the estimates are based on a regression assumption that always omits vital variables. Also, empirically derived gravity models do not consider the size of economies and their effects on trade between states and provinces. Secondly, it implies that economists can draw comparative statistics inferences from their estimates using the theoretically derived gravity model. This is because the model has theoretical foundations that economists can apply consistently, in their comparisons. On the other hand, regression model authors are unable to draw comparative statistic inferences from their estimates because there are no theories or theoretical foundations to apply consistently. For instance, taking the existing gravity theory seriously provides a different model to estimate with a useful interpretation. Economists can determine whether there is an invariance of trade to uniform decreases, using the theoretically derived gravity model. Sadly, the empirically based gravity model estimation cannot find out whether trade becomes insensitive to distance, over time. Another implication of the superiority of the theoretically derived gravity model is provision of unbiased estimation results in analysis of impacts of elements of bilateral trade (Deardorff, 1998). There is no omission of variables when using the theoretically derived gravity model. Therefore, economists can conduct a comparative statistics exercise using the theoretically derived gravity model given that this is the purpose of estimating gravity equations. The estimation of effects of distance and output that economists make, using the theoretically derived gravity model is economically and statistically significant. This is because the theory in the model explains most of the variation in international trade (García, Tortosa, & Fernandez, 2009). Economists can be able to determine whether trade barriers reduce adjusted trade size between large countries more than between small countries. Trade between regions is determined by comparative trade barriers. Theory suggests that trade between two regions depends on the bilateral barrier between them, comparative to average trade barriers that both regions face with their trading partners. The theoretically derived model can be used by economists to establish whether trade barriers raise size altered trade within small countries more than within large countries. The model is also vital in establishing whether trade barriers raise the proportion of size regulated trade within a country, compared to size altered trade between countries. The superiority of the theoretically derived gravity model implies that a comparative analysis of the effects of trade barriers can be derived under a few assumptions using the model. Economists can be able to establish whether international trade can be generated without natural or acquired comparative advantages of different countries. The theoretically derived gravity model is ideal for carrying out a reliable comparative analysis. Economists can easily incorporate sources of international trade between nations, comparative to differences in factor endowment into the framework. A common implication is that the theoretically derived gravity model does not only specify the bilateral distance between countries, omitting a potentially imperative explanatory variable. It incorporates the bilateral distance (Brun, Carrere, Guillaumon, & Melo, 2003). Transport costs between two countries, in comparison with a measure of overall transport costs is an imperative variable for analysis. Finally, the theoretically gravity model incorporates trade dynamics in analysis. This implies that there will be no incorrect inferences made because of ignoring the effects of lagged trade on current trade. Empirically based gravity models the effects of ignoring trade dynamics (Bun & Klaassen, 2002). This leads to the formulation of incorrect inference because lagged trade affects current trade. Economists can use the theoretically derived gravity model to compare the impact of unobserved time invariant factors that affect both lagged and current trade. The theoretically derived gravity model does not led to the formation of bias estimation in comparative statistics. It includes zero observations, leading to unbiased estimations of the impact of distance over time because of the changing composition of trade. Empirically based gravity models could lead to bias estimates, in comparative statistics because they exclude zero observations, in their analysis. They also omit multilateral trade resistance. This implies that the use of the theoretically derived gravity model in comparative statistics is easy to implement. It has a superior empirical performance because data requirements for Econometric Analysis of Trade Flows are low. This is because the model relies on extensively available information. The estimation procedure used by the theoretically derived gravity model is straight forward (Medvedev, 2006). In conclusion, the theoretically derived gravity model provides reliable estimates for comparative statistics. These estimates may be applied in economic analysis of trade flows. This is because the framework incorporates theoretical foundations in the development of models for comparative statistic analysis. References Anderson, J. E., & Wincoop, E. V. 2001. Gravity with Gravitas: A Solution to the Border Puzzle. Retrieved August 28, 2012, from https://www2.bc.edu/~anderson/BorderEffects.pdf Anderson, J., & Wincoop, E. V. 2004, April 30. Trade Costs. Retrieved August 28, 2012, from http://fmwww.bc.edu/ec-p/wp593.pdf Baier, S. L., & Bergstrand, J. H. 2005, February 15. Do Free Trade Agreements Actually Increase Members Iinternational Trade? Retrieved August 28, 2012, from http://www.frbatlanta.org/filelegacydocs/wp0503.pdf Baldwin, R., & Taglioni, D. 2006, September. Gravity for Dummies and Dummies for Gravity Equations. Retrieved August 28, 2012, from http://www.nber.org/papers/w12516 Brun, J. F., Carrere, C., Guillaumon, P., & Melo, J. d. 2003, March 22. Has Distance Died? Evidence from a panel Gravity Model. Retrieved August 28, 2012, from http://www.cepr.org/meets/wkcn/2/2339/papers/deMelo.pdf Bun, M. J., & Klaassen, F. J. 2002, March 27. The Importance of dynamics in Panel Gravity Models of Trade. Retrieved August 28, 2012, from http://dare.uva.nl/document/334911 Deardorff, A. 1998, January. Determinants of Bilateral Trade: Does Gravity Work in a Neoclassical World? Retrieved August 28, 2012, from http://www.nber.org/chapters/c7818.pdf García, F. P., Tortosa, E., & Fernandez, I. 2009. The Distance Puzzle Revisited: A New Interpretation Based on Geographic Neutrality. Bilbao: University of Valencia Press. Leamer, E. E., & Levinsohn, J. (1994, June). International Trade Theory: The Evidence. Retrieved August 28, 2012, from http://ideas.repec.org/p/nbr/nberwo/4940.html McCallum, J. 1995, June. National Borders Matter: Canada-U.S. Regional Trade Patterns. Retrieved August 28, 2012, from http://www.jstor.org/stable/2118191 Medvedev, D. 2006. Preferential Trade Agreements and Their Role in World Trade. Washington, D.C: World Bank Publications. Rose, A. K. 2004, March. Do We Really Know That the WTO Increases Trade? Retrieved August 28, 2012, from http://www.jstor.org/stable/3592771 Subramanian, A., & Wei, S.-J. 2006, July 25. The WTO promotes Trade, Strongly but Unevenly. Retrieved August 28, 2012, from http://www.imf.org/external/pubs/ft/wp/2003/wp03185.pdf Read More