Analysis of Trade Effect in Post-Tpp Era: Based on Gravity Model and Gtap Model

The basic idea of the gravitational model is derived from Newton’s law of universal gravitation found in physics, that is, there is a certain attraction between an object and another object in the universe, and the magnitude of gravity between the two is determined by the mass between the two objects. Distance decision. Specifically, the magnitude of gravity changes positively with the respective masses of the two objects, and the distance between the two objects changes inversely, and is expressed as follows: (1)F=GM1M2/D2F = G{M_1}{M_2}/{D^2} F in the formula represents the magnitude of gravity between two objects, M₁ and M₂ represent the mass of two objects, D is the distance between two objects and G represents gravitation, which is a constant, indicating that when the mass of two objects is exactly equal to the unit mass and the distance between the two objects is exactly equal to the unit distance, the mutual attraction between the two.

In 1962, Tinbergen first used Newton’s gravity model for international trade research and added parameters related to international trade [15], changing the original gravity model to the following form.

This model became the original model of gravity model introduced into the field of economics. Later, domestic and foreign scholars also used this model to analyse the trade effects of various regional free trade arrangements. In the formula, the physics formula of the gravity model is fine-tuned. The economic meaning of replacing F with X_ij is the trade gravity between two economies or countries, that is, the size of trade volume, and Y_i and Y_j replace two objects. The quality, meaning the economic size of the two economies, is usually expressed in terms of the GDP of the economy. The meaning of D_ij is similar to D in (1), which means the distance between the two economies, usually between the capitals of the two countries. The distance indicates that the distance between the important economic cities of the two countries is also expressed in practice, which is mainly related to the needs of the research objects. And β₁, β₂ and β₃ respectively indicate the degree of influence of economic scale and distance variables of two economies on the change of trade volume, and A is a trade constant, meaning similar to G in (1), which is also a constant, indicating the economy of two economies. The scale is exactly one unit size, and the trade volume between the two is just one unit. Later, scholars continued to perfect the basic gravity model of Tinbergen. The refined model is as follows: (3)Xij=C(Yi)β1(Yj)β2/(1+eDij)γ{X_{ij}} = C{\left( {{Y_i}} \right)^{{\beta _1}}}{\left( {{Y_j}} \right)^{{\beta _2}}}/{\left( {1 + e{D_{ij}}} \right)^\gamma }

The improved model is similar in economics to (2), but adding the coefficients of the distance variable will make the operation more accurate. To facilitate the solution, the nonlinear equation is linearised by logarithm. When the random error term is added, the following form is obtained, which also becomes the basic form commonly used in the future gravity model and the basis for studying international trade-related issues: (4)LnXij=LnC+β1Ln(Yi)+β2Ln(Yj)+γLn(Dij)+μijLn{X_{ij}} = LnC + {\beta _1}Ln({Y_i}) + {\beta _2}Ln({Y_j}) + \gamma Ln({D_{ij}}) + {\mu _{ij}}

In 1966, H. Linnemann proposed that the parameters of (4) basic forms are too simplified, which is not conducive to the analysis of specific problems in international trade. It is suggested to add demographic variables to the model. However, many scholars have since opposed the inclusion of models, such as McCallum, Anderson, Wei Shangjin, and many more. They believe that the population does not determine trade as compared with important factors such as the output level that determines the volume of trade. The necessary factors are considered for change, so whether population variables should be included in the gravity model has always been the object of scholars’ arguments. Instead, people usually add per capita GDP to the model. Because this article examines the determinants of the trade relationship between the United States and TPP member states, it is logically considered that the trade volume is mainly determined by the market size of TPP member countries, and there are many determinants of market size, but it does not include population factors, so the following. There is also a view that there is no absolute direct correlation between demographic factors and trade volume, and population factors are not included in the model. In addition, Linnemann (1966) also included preferential trade agreement (PTA) as a dummy variable into the equation, mainly to investigate the impact of regional trade liberalisation on bilateral trade flows. At this point, the gravity model includes quantitative variables like GDP and distance and also includes regional economic integration, openness, whether to use the same language, whether the common boundary is used and the virtual variables that are expressed in a non-formal form. The form of the model is also based on actual needs. Continuous improvement and complementation, for example, of the equations that are established by Franke (1997) and Freund (2000) are given as follows: (5)LnTij=a+β1Ln(GDPiGDPj)+β2Ln(Distij)+β3Ln(PCiPCj)+β4Ln|PCi−PCj|+α1ADFij+α2RTAij+α3Openij+μij\matrix{ {Ln{T_{ij}} = a + {\beta _1}Ln(GD{P_i}GD{P_j}) + {\beta _2}Ln(Dis{t_{ij}}) + {\beta _3}Ln(P{C_i}P{C_j}) + {\beta _4}Ln\left| {P{C_i} - P{C_j}} \right| + {\alpha _1}AD{F_{ij}} + {\alpha _2}RT{A_{ij}}} \hfill \cr { + {\alpha _3}Ope{n_{ij}} + {\mu _{ij}}} \hfill \cr }

Among them, in Eq. (5), Tij represents the total trade between economies i and j; Distij represents the measure of distance; PCi is the per capita GDP of country i, PC_j is the GDP per capita of j; PCiPCj is the GDP per capita of both countries i and j Product, representing the importance of “wealth” (corresponding to scale) as a determinant of trade; |PC_i-PC_j| is the absolute value of the difference in GDP per capita, representing the difference between economies; α and β are parameters; ADFij represents the existence of a boundary; RTAij represents the existence of a regional trade agreement (if the economies i and j are members of the regional trade agreement being discussed, the value is 1); openij represents the degree of openness of the member states (if the countries i and j are positive). The member of the regional trade agreement being discussed has a value of 1); μ_ij is the error term. This article also adds relevant dummy variables based on the extended gravity model and strives to establish a model that conforms to the actual trade relationship between the US and TPP members.

The application of the gravity model in this article is mainly to determine the determinants of the trade volume between the United States and other TPP members. Therefore, all the factors that may affect the volume of US trade when building a model must be considered.

(1) Economic factors

The first consideration in the model is the GDP of the TPP member countries that trade with the United States. As a measure of the economic size of the trading partner countries, it is a necessary parameter in the gravity model, so it is directly incorporated into the model according to the requirements of the classical form. Another necessary parameter in the gravity model is the distance factor. As the trade cost, it will also be directly included in the model. It is logically believed that the greater the distance between the two countries, the greater the transportation cost, and thus the greater the trade cost. Measuring the cost of transportation through the distance between the United States and TPP members is an important factor in determining the volume of trade with the United States. Besides the above quantitative factors, this article also includes the degree of trade liberalisation in economic factors as a dummy variable, that is, considering which countries in the US and TPP countries have previously signed a free trade agreement because generally, the two countries that sign the free trade agreement will promote the trade volume between the two parties. To verify the determinants of the trade relationship between the United States and the TPP member states, other trade agreements other than the TPP signed by the United States will naturally become an important consideration factor.

On the basis of the aforementioned analysis, an extended gravity model is established, which is shaped as follows: (6)LnT0j=α0+α1LnY0+α2LnYj+α3LnD0j+α4UCFTA+α5UAFTA+α6UPFTA+α7USFTA+α8NAFTA+μij\matrix{ {LnT{0_j} = {\alpha _0} + {\alpha _1}Ln{Y_0} + {\alpha _2}Ln{Y_j} + {\alpha _3}LnD{0_j} + {\alpha _4}UCFTA + {\alpha _5}UAFTA + {\alpha _6}UPFTA} \hfill \cr { + {\alpha _7}USFTA + {\alpha _8}NAFTA + {\mu _{ij}}} \hfill \cr }

Among them, T0_j represents the bilateral trade volume between the United States and various TPP member countries, Y0 represents the US GDP, Y_j represents the GDP of the US trading partners, D0_j represents the distance between the United States and the trading partners, UCTTA, UAFTA, UPFTA, USFTA, NAFTA is a dummy variable representing the United States–Chile Free Trade Agreement, the US–Australia Free Trade Agreement, the US–Peru Free Trade Agreement, the US–Singapore Free Trade Agreement, and the North American Free Trade Agreement, which are signed in the year of non-effectiveness. The year in effect is 1.

(2) Political and military factors

Besides economic factors, because the US-led TPP may be for political and military purposes, that is, to maintain stability and absolute dominance in the Asia-Pacific region, it may give priority to whether the United States has signed a military agreement with it when it chooses a country of trade. The alliance agreement, because the United States has a traditional military ally in the Asia-Pacific region, has also considered it as an important dummy variable. Because political and military factors have many considerations, it is difficult to use quantitative parameters. Therefore, this article only includes the dummy variable of military–political alliance as the main indicator of political and military factors. At present, the US military and political allies in the Asia-Pacific region mainly include Japan, ASEAN countries, Australia and other countries, so they join the three dummy variables of Union1, Union2 and Union3, representing the US-Japan military alliance, the US–ASEAN partnership, and the US-Australia. The alliance, the effective year is 1, and the remaining years are 0, the final gravitation model is obtained: (7)LnT0j=α0+α1LnY0+α2LnYj+α3LnD0j+α4UCFTA+α5UAFTA+α6UPFTA+ α7USFTA+α8NAFTA+Union1+Union2+Union3+μij\matrix{ {LnT{0_j} = {\alpha _0} + {\alpha _1}Ln{Y_0} + {\alpha _2}Ln{Y_j} + {\alpha _3}LnD{0_j} + {\alpha _4}UCFTA + {\alpha _5}UAFTA + {\alpha _6}UPFTA} \hfill \cr { + {\alpha _7}USFTA + {\alpha _8}NAFTA + Unio{n_1} + Unio{n_2} + Unio{n_3} + {\mu _{ij}}} \hfill \cr }

Therefore, (7) is used as the gravitational model, in this article, to analyse the factors affecting the bilateral trade flows between the United States and TPP member countries, and to introduce the future development trend of the United States to further lead the TPP.