The effect of regional trade agreement to trade flow evidence of trade creation and trade diversion of asean – japan free trade area

INTRODUCTION

Problem statement

Regional trade agreements (RTAs) have gained significant traction in the international economic landscape, particularly following the Doha Round of the GATT/WTO Defined by the WTO, RTAs encompass free trade agreements (FTAs) and customs unions (CUs), involving negotiations between two or more parties to reduce customs barriers such as tariffs and quotas Since the early 1990s, the prevalence of RTAs has surged, with the WTO reporting 625 notifications of RTAs and 419 currently in force as of February 2016 The Association of Southeast Asian Nations (ASEAN) exemplifies a successful model of regionalism, progressively enhancing cooperation and integration within the global economy Meanwhile, Japan's economic contribution has seen a decline, dropping from 17% of the world economy in 2005 to just 6% in 2015, according to the IMF.

Japan's economic performance significantly impacts the entire region, as it ranks among the top three trading partners of ASEAN economies, particularly benefiting Indonesia and the Philippines.

Before Japan's integration into ASEAN regional economies, it played a crucial role in regional development, with 25% of ASEAN's total import and export values linked to Japan in the 1970s The lower costs of materials and labor made ASEAN markets attractive for Japanese capital investment, generating jobs and increasing wages Additionally, Japanese companies introduced high technologies and skilled training, creating valuable learning and transfer opportunities in the region during the 1980s and 1990s This growing business integration highlighted the need for strengthened linkages between ASEAN and Japan, underscoring the importance of discussing a regional agreement.

Since 2003, Japan and the 10 ASEAN countries have established the ASEAN-Japan Free Trade Agreement (AJCEP), a comprehensive economic partnership aimed at enhancing trade relations and economic cooperation in the region.

2008, the last official round was finalized, an agreement signed among Asian countries, included: Brunei Darussalam, Cambodia, Indonesia, Laos PRD, Malaysia, Myanmar, Philippines,

Singapore, Thailand, Vietnam, and Japan are promoting multilateral trade by reducing tariffs through a Free Trade Agreement (FTA) This initiative aims to foster free trade within ASEAN and Japan, strengthen economic integration among Asian countries, enhance their global market presence, ensure transparency in trading procedures, and promote sustainability in the economic sector This presents significant opportunities for Japan's high-tech and modern industries, such as automotive and electronics, to penetrate ASEAN markets and encourage the establishment of assembly lines in the region for Japanese companies.

Since the implementation of the trade agreement in 2013, two-way trading volume between Japan and ASEAN has surged to $229 billion, a significant increase from $128 billion in 2000 In 2014, Japan's imports from ASEAN accounted for 14% of its total import value, with Thailand ($22.5 billion), Indonesia ($32.2 billion), and Malaysia ($29.6 billion) being the top exporters Key exports from ASEAN to Japan include food, manufactured goods, textiles, and raw materials In contrast, Japan primarily exports machinery, transportation equipment, chemicals, and advanced technology products to ASEAN countries Notably, in 2014, 47% of Japanese cars, 80% of trucks, and 85% of buses were consumed in ASEAN markets, highlighting the region's importance to Japan's automotive industry.

Research objectives

This study aims to investigate the impact of the ASEAN-Japan Comprehensive Economic Partnership (AJCEP) on ASEAN economies, specifically focusing on trade creation and trade diversion effects using total export data.

The second research objective focuses on analyzing the impact of the AJCEP agreement on specific sub-categories, including food products, agricultural goods, manufactured items, transportation machinery and equipment, as well as clothing, accessories, and textiles.

Research questions

Numerous studies indicate that Regional Trade Agreements (RTAs) do not consistently ensure positive outcomes for member countries' integration into the global market; in fact, they often lead to negative consequences This study seeks to explore these issues further.

- How the trade creation and trade diversion in general total export have been caused by the free trade agreement which was signed by AJCEP to ASEAN member countries?

The AJCEP free trade agreement has significantly influenced trade creation and trade diversion among ASEAN member countries across five key sub-categories: food products, agricultural products, manufactured goods, machinery and transportation equipment, and textiles and clothing By reducing tariffs and trade barriers, the agreement has facilitated increased trade flows in food and agricultural sectors, while also promoting the exchange of manufactured products and machinery However, it has also led to trade diversion, as some countries may shift imports from non-member nations to those within the agreement, impacting local industries Overall, the AJCEP fosters economic integration and enhances competitiveness within the ASEAN region.

Research scope

To estimate the effect of AJCEP, we employ a panel data set will be collected with period from

Between 2000 and 2015, a total of 5,920 observations were recorded involving nine ASEAN countries: Brunei Darussalam, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, the Philippines, Singapore, Thailand, and Vietnam In 2015, Japan's fifteen largest trading partners included the United States, China, South Korea, Hong Kong SAR, Australia, Saudi Arabia, the United Arab Emirates, the Russian Federation, Switzerland, New Zealand, the United Kingdom, Germany, Mexico, the Netherlands, and Japan itself.

Recent research highlights the swift evolution of the impact of Regional Trade Agreements (RTAs) on investment, both theoretically and empirically However, many studies tend to concentrate on broad inquiries, such as whether RTAs influence trade flow or lead to trade creation and diversion This focus reveals two significant issues that have been prevalent in previous research.

The gravity model faces estimation challenges due to heteroscedasticity and frequent zero trade observations, which complicate the selection of appropriate techniques and may lead to biased results To address this, we utilize two distinct trade flow datasets: an aggregated dataset for analyzing total bilateral export flows and a disaggregated dataset for assessing the impact of AJCEP across five specific subcategories: agriculture, manufacturing, chemical industry, machinery, and textiles This dual approach allows for a comprehensive analysis of AJCEP's overall effects as well as its influence on particular commodities.

Thesis structure

This paper is structured as follows: Chapter 2 reviews literature on trade theories related to international trade flows, focusing on the theoretical and empirical support for the gravity model and examining the impact of AJCEP on ASEAN member countries and Japan, while addressing the common issue of zero trade data Chapter 3 outlines the methodology, including model construction, estimation methods, and the data scope utilized in the study Chapter 4 presents the results and findings derived from the regression model, and Chapter 5 concludes with a summary of the thesis results, recommendations, and limitations of the study.

LITERATURE REVIEW

This chapter provides a summary of key trade theories commonly utilized in international trade, followed by an examination of the theoretical and empirical backing for the gravity model in trade Additionally, it explores literature on zero trade data and the development of estimation techniques employed in previous studies.

International trade offers significant benefits to countries, as explained by various theories, including Adam Smith’s Absolute Productivity Advantage Model Smith posited that when labor and production factors are fixed and fully utilized, with constant technology and zero transportation costs, countries should focus on exporting domestically effective products while importing those that are more effective from abroad (Howse and Trebicook, 1995) When a country possesses an absolute advantage in producing two goods, David Ricardo's theory of comparative advantage comes into play This theory suggests that countries should produce goods with lower opportunity costs, allowing them to specialize according to their comparative advantages Consequently, both nations can benefit from specialization and engage in profitable international trade.

The law of comparative advantage emphasizes labor as the sole factor in production and trade, while the Heckscher-Ohlin model highlights the significance of resource abundance differences, including labor, capital, and land, across countries Simply noting the quantity of resources like labor or land is insufficient; thus, the concepts of labor-intensive and land-intensive production have been introduced Consequently, countries tend to specialize in manufacturing products that utilize their most abundant factors of production.

2.2 Trade creation and trade diversion

Before Viner's 1950 analysis, studies generally suggested that tariffs diminished welfare, leading to the belief that customs unions or free trade agreements (FTAs) would enhance welfare Viner introduced the concepts of trade creation and trade diversion effects associated with FTAs, arguing that their impact on welfare is not universally positive The overall effect of an FTA on welfare is contingent upon the balance between trade creation and trade diversion; if trade creation outweighs trade diversion, the FTA is likely to improve welfare.

When a Free Trade Agreement (FTA) is implemented, it typically eliminates tariffs and trade incentive policies, fostering trade flows that previously did not exist This arrangement enables member countries to focus on their comparative advantages, leading to benefits from economies of scale Consequently, the resulting trade creation enhances the national welfare of the participating countries.

Trade agreements and customs unions often lead to trade diversion, where goods are imported from less efficient exporters within the union instead of more efficient ones outside it, due to the removal of tariffs between member countries while non-members still face common tariffs This results in higher costs for more efficient countries, diminishing their competitive advantage Consequently, trade flows shift towards member countries, benefiting them from trade diversion, but negatively impacting non-member countries economically Ultimately, the removal or reduction of tariffs allows higher-cost producers to access the market, leading to a loss in overall welfare due to inefficient trade patterns.

Figure 1 shows the welfare effects of joining a free trade agreement � and � are denoted for domestic demand and domestic supply of a specific product � of a country respectively.

� � and � � are represented for exporting supply of product � from intra-bloc member countries and extra-blocs non-member countries.

Before the integration of a free trade agreement, the supply curves for intra-bloc and extra-bloc are represented as $ S_{intra} $ and $ S_{extra} $ respectively It is assumed that non-member countries offer product $ X $ at a lower price than member countries, indicating that $ S_{extra} $ is positioned below $ S_{intra} $.

� lies under � � + � graphically The difference between � 0 − � 0 is country import demand from non-members.

Following the establishment of a free trade agreement, the supply curve remains unchanged due to the continued application of tariffs on non-member countries However, tariffs are no longer applied to imports from member countries As a result, the equilibrium price of product $ P $ in the country will be $ P_1 $, with the import demand shifting from non-member to member countries, reflected in the difference $ P_1 - P_1 $ Additionally, the domestic consumer surplus is represented by the area sum, highlighting the benefits of the agreement.

As domestic manufacturers experience a decline in surplus, the government's tax revenue diminishes due to the implementation of a free trade agreement, which eliminates taxes on imports from member countries Consequently, the overall impact of trade creation resulting from the free trade agreement is represented by the combined value of areas A and B.

Trade diversion occurs when countries shift their imports from lower-cost producers outside their trade bloc to higher-cost producers within the bloc, represented by the designated area The overall impact of a free trade agreement is assessed by comparing the levels of trade creation and trade diversion If the benefits of trade creation surpass the drawbacks of trade diversion, the welfare of the country improves as a result of the agreement Conversely, if trade diversion outweighs trade creation, it indicates a decline in national welfare due to the free trade agreement.

Figure 1: Trade creation and trade diversion

2.3 The gravity model in international trade

2.3.1 The origin of gravity model

The gravity model serves as a fundamental tool for analyzing international trade flows, drawing its principles from Isaac Newton's law of universal gravitation established in 1967 This law posits that two points exert an attractive force on each other, which is directly proportional to the product of their masses and inversely proportional to the square of the distance separating them.

� �� is the gravity force between two of masses

� � , � � are the masses of the first and second point respectively

� 2 is the distance from fist point center to the second point center in square

The gravitational constant, denoted as G, has a fixed value of 6.674 x 10^-11 N(m/kg)^2 In 1962, Tinbergen pioneered the application of the gravity model, based on Newton's law of gravitation, to examine international trade flows In this trade model, the bilateral trade volume between two countries is utilized to represent the gravitational force, reflecting their economic sizes.

� � , � � have been used to replace for the masses of � � , � � respectively Generally, the gravity formulation has been established in the following form:

Where �, �, � may take the value different to 1 They depend on the elasticity of economic sizes

� � of exporting country, importing country and distance respectively In case, � = � = 1 and �

Economic indicators such as GDP, GNP, real GDP, real GNP, income per capita, and population are crucial metrics that reflect the supply and demand dynamics within a country, ultimately influencing its trade volume These variables are analogous to Newton's equation, highlighting their fundamental role in understanding economic performance.

Distance, defined as the geographical separation between two economic hubs or capitals measured in land miles, plays a crucial role in trade dynamics According to Tinbergen, distance serves not only as a representation of physical space but also encapsulates various market factors that impact trade volume, including transportation costs, transit expenses, communication exchange costs, and cultural factors.

In economic regression, the simplest gravity model is typically estimated using Ordinary Least Squares (OLS) by applying a logarithmic transformation to the equation and incorporating an error term The resulting coefficients are interpreted as elasticities due to the double log form of the regression.

(3) According to the explanation above, the coefficients will be interpreted as follow:

RESEARCH METHODOLOGY

This chapter examines the gravity model in international trade, transitioning from multiplicative to logarithmic form by incorporating additional variables that influence bilateral trade It specifically investigates the impact of the ASEAN-Japan Comprehensive Economic Partnership (AJCEP) on both ASEAN countries and Japan Various estimation techniques, including Ordinary Least Squares (OLS), Fixed Effect Model (FEM), Random Effect Model (REM), and Hausman-Taylor estimator, are utilized for analysis Additionally, the chapter provides an overview of the data scope and sources used in the study.

3.1 Model specification and validity testing

Starting with the multiplicative equation of gravity model in international trade:

According to the gravity model of international trade, the total bilateral export volume between two countries is influenced by their GDPs, populations, and the distance between them Additionally, a set of dummy variables, represented by "� ��," plays a crucial role in either promoting or hindering bilateral trade flows, including factors such as borders.

Language, Colony, Land-lockedness, Free trade agreement.

For estimation target as well as time dimension plus, we change model (7) into log-liner format which is given as below:

Trade flow (� �� ): in the gravity model, we employ variable trade flow by export volume from exporting country � to importing country � at time t at current US$.

Gross Domestic Product (GDP) at current US dollars is a key independent variable sourced from the World Trade Organization (WTO) database This variable represents the total value of all final goods and services produced in a country over a specific time period, measured in current US dollars In the context of the gravity model, it plays a crucial role in understanding trade dynamics.

According to utility theory, an increase in a country's income and output leads to higher consumer demand for goods and services, which in turn boosts production and exports Nellis and Parker (2004) suggest that a country's GDP positively correlates with the total volume of imports and exports, indicating that higher income and purchasing power enhance trade activity However, Basat (2002) notes that this positive relationship is primarily observed in middle-development countries, with no evidence supporting it in low or high-development countries.

The population is the second independent variable, divided to two population variables,

The population of the exporting country is denoted as $ P_e $ and the population of the importing country as $ P_i $ at time $ t $, measured in millions It is anticipated that countries with larger populations will exhibit higher demand for both imports and exports However, Aiken (1973) suggested that in populous nations, the ratio of domestic market demand to foreign market demand exceeds one, implying that countries with smaller populations may achieve greater export volumes Consequently, Oguledo and Macphee (1994) indicate that the expected signs for $ P_e $ and $ P_i $ will vary based on each country’s level of integration.

Weighted distance, introduced by Mayer and Zignago (2005), serves as the third independent variable in their analysis This concept builds on the earlier work of Head and Mayer (2002) and measures the geographic distance between two countries by considering the distance between their largest cities Additionally, this measurement incorporates a weighting factor based on the population ratio of each city to the overall population of its respective country.

Using this method over simple geodesic distance, which relies solely on longitudes and latitudes, helps prevent the overestimation or underestimation of border effects For instance, when analyzing trading volume between Vietnam and China, factors such as domestic transportation costs within China, international transportation costs between China and Vietnam, and the population in relevant cities significantly influence trade volume The foundational formula for calculating the weighted distance from country A to country B was developed by Head and Mayer (2002).

� �� is the weighted-distance between two countries � and �

�� , �� are population agglomeration of cities �, � belongs to country � and � respectively

�� , �� are the total population of country � and � respectively

� �� is the geographic bilateral distance between two city �, �

Bakman, Garretsen, and Marrewijk (2001) highlight that increased distance leads to higher transportation costs and greater cultural differences, which in turn elevate trade costs This indicates a negative correlation between distance and trade flow.

Language is the first dummy variable, used as a measurement tool to compare culture facture differences According to the theoretical support of Linnermann (1966), Hacker and Johansson

In 2001, research indicated that trade volume between two countries is significantly affected by the presence of a common language, which serves to reduce communication barriers When countries share a language, they can communicate more easily, leading to lower transaction costs Consequently, a binary variable is utilized: it equals 1 if the countries share a language and 0 if they do not, with the expectation that the coefficient will yield a positive sign.

Countries that share a common border experience reduced transportation costs, which facilitates increased trade flow (Bakman, Garretsen, and Marrewijk, 2001) Consequently, a binary variable is utilized, assigning a value of 1 to pairs of countries that share a border and 0 to those that do not It is anticipated that this variable will have a positive coefficient.

When two countries have a colonial history with each other or a shared colonizer, a specific dummy variable is utilized to indicate their relationship This variable is assigned a value of 1 if the countries share a common border and 0 if they do not Consequently, it is anticipated that the coefficient associated with this variable will have a positive sign.

In international trade analysis, a dummy variable is used to indicate whether at least one country in a pair is landlocked, with a value of 1 for landlocked countries and 0 otherwise Research has shown that landlocked countries face higher import prices and reduced export revenues due to increased costs in intermediate export services (Mackellar, Worgotter, and Worz, 2002; Stone, 2001) Consequently, the expected outcome for the coefficient associated with this variable is a negative sign.

(�� ) is a dummy variable are extended three different set of binary variables, included

The ASEAN-Japan Free Trade Agreement (AJCEP) significantly influences trade dynamics among member countries The variable ��_1 indicates whether both countries, designated as �� and ��, are AJCEP members post-2008, with a value of 1 for membership and 0 otherwise Meanwhile, ��_2 represents exporters that are AJCEP members in year t, while the importing country �� is not, assigning a value of 1 for membership and 0 otherwise Lastly, ��_3 reflects cases where the exporter is from an AJCEP non-member country exporting to an AJCEP member country, receiving a value of 1 for such transactions and 0 otherwise.

A positive and statistically significant coefficient for ��_1 in the regression results indicates that a trade creation effect is generated, suggesting that intra-regional trade flow increases when a free trade agreement (FTA) is in effect Similarly, a positive and statistically significant coefficient for ��_2 signifies that the FTA has also fostered trade creation in terms of exports, indicating a shift in export activities.

The analysis of AJCEP member countries reveals a trade diversion effect, as indicated by a statistically significant negative coefficient for exports to non-member AJCEP countries Conversely, the regression results demonstrate a positive and statistically significant coefficient for imports from extra-bloc to intra-bloc AJCEP, suggesting a trade creation effect This indicates that AJCEP has facilitated an increase in export flows from non-member countries to member countries However, if the coefficient for imports were to show a negative and statistically significant value, it would imply a trade diversion effect in relation to imports.

RESEARCH FINDINGS AND DISCUSSION

This section comprises two sub-sections: the first focuses on the statistics of variables and their economic implications, while the second presents econometric results and their interpretations Additionally, the second sub-section will include tests to identify the most suitable estimated model and eliminate any inappropriate models.

Variables summary statistics of panel data are presented in Table 2

Table 2 presents data from 24 countries over a 15-year period (2000-2015), encompassing 5,920 observations Total export value constitutes 0.34% of the overall gross domestic product Some countries report a minimum export value of 0, which may stem from missing data, rounding errors, or a lack of inter-country exports Manufactured goods dominate the total export trading value, accounting for approximately 77.80%, with chemical products and machinery and transportation equipment comprising 8.64% and 48.72%, respectively Agricultural products represent 6.94% of total exports.

Table 2: Descriptive statistics of variables

Machinery and transport equipment 5,920 2.133e+09 7.129e+09 0 export

Cloth and textiles products export 5,920 1.909e+08 1.208e+09 0

The dummy variable ��_1 is set to 1 when both the importing and exporting countries are part of the AJCEP free trade agreement established in 2008, with each country's effective date adjusted as detailed in Table 10 Descriptive statistics for other variables when ��_1 equals 1 are presented in Table 3.

Table 3: Descriptive statistics of variables if pair of countries belongs to AJCEP from 2008

Gross domestics products (Exporting country) 596 6.71E+11 1.54E+12 0 5.96E+12

Gross domestics products (Importing country) 596 6.70E+11 1.53E+12 0 5.96E+12

Machinery and transport equipment export 596 1.80E+09 3.75E+09 0 2.51E+10

Cloth and textiles products export 596 8.29E+07 2.47E+08 0 3.00E+09

The dummy variable ��_2 is set to 1 when the exporting country is part of the AJCEP, while the importing country is not included in the AJCEP free trade agreement since 2008 Table 4 presents the descriptive statistics for other variables when ��_2 equals 1.

Table 4: Descriptive statistics of variables if exporting country is belong to AJCEP and importing countries is not belongs to AJCEP from 2008

Gross domestics products (Exporting country) 952 6.72E+11 1.54E+12 0 5.96E+12

Gross domestics products (Importing country) 952 3.06E+12 4.19E+12 2.14E+11 1.79E+13

The dummy variable ��_3 is assigned a value of 1 when the exporting country is not a member of the AJCEP, while the importing country is part of the AJCEP free trade agreement established in 2008 Table 5 presents the descriptive statistics for other variables when ��_3 equals 1.

Table 5: Descriptive statistics of variables if exporting country is not belong to AJCEP and importing countries is belongs to AJCEP from 2008

Gross domestics products (Exporting country) 966 3.03E+12 4.20E+12 1.94E+11 1.90E+13

Gross domestics products (Importing country) 966 7.10E+11 1.66E+12 0 1.79E+13

Multicollinearity can significantly affect regression outcomes, making it essential to detect its presence among variables This can be achieved through Pearson correlation and the calculation of Variance Inflation Factors (VIFs) As illustrated in Table 6, the correlation between the variables in this study is presented.

Testing results indicate a positive correlation between export variables and several factors: GDP of exporting countries (0.3068), GDP of importing countries (0.346), population of exporting countries (0.2532), population of importing countries (0.2251), common border (0.0008), and common language (0.1272) This suggests that wealthier countries are more likely to import goods, while a shared language facilitates communication and reduces transaction costs Additionally, countries sharing a border tend to have similar cultures and lower transportation costs, which enhances export trading flows.

The analysis indicates that factors such as colonial ties, geographical distance, and landlocked status negatively impact export levels Specifically, a correlation of -0.0218 for colonial relationships, 0.0617 for distance, and -0.086 for landlocked countries suggests that when two countries share a colonial history, are located far apart, or if one is landlocked, their export activities are likely to decline.

Regarding FTA variables, ��_2 and ��_3 show the positive relation with export while

The findings indicate a negative relationship, suggesting that the AJCEP facilitates trade between its member countries and Japan, but has not yet succeeded in enhancing trade among its own members.

Population import country Export Colony Distance Border

The results are divided into two sections Initially, we review the econometric findings by employing various estimation methods, including OLS, Fixed Effect, Random Effect, and the Hausman-Taylor estimator, to identify the most suitable estimator Subsequently, we will discuss the parameters derived from these estimations.

Figure 2: Testing residual against total in level (OLS estimation)

Previous studies indicate that Ordinary Least Squares (OLS) often incorporates heterogeneity bias, leading to biased estimations if unaddressed This bias arises when model variables are correlated, affecting the accuracy of results Figure 2 illustrates this issue, showing a negative residual at low total export levels, which suggests that exports are overestimated.

To assess whether the Fixed effect or Random effect model is superior, we will conduct a Hausman test, which tests the null hypothesis that the coefficients from the random effect estimator are equivalent to those from the fixed effect estimator If we accept the null hypothesis, it indicates that the Random effect model should be utilized The outcomes of the Hausman test are presented in Table 7.

Based on the testing results, a P-value of less than 0.05 indicates strong evidence against the null hypothesis, leading us to conclude that the fixed effect model is more appropriate than the random effect model.

Fixed b = consistent under Ho and Ha; obtained from xtreg Random (B) = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic chi2(7) = (b-B)'[(V_b-V_B)^(-1)](b-B)

= 143.76 Prob>chi2 = 0.0000 (V_b-V_B is not positive definite)

Table 8 shows the overview about numeric regression result Each column presents each type of estimator: OLS, fixed effect model, random effect model, Hausman-Taylor respectively.

Table 8: OLS, FEM, REM and Hausman-Taylor regression result for total export

(4) Hausman- Taylor Log GDP exporting country

Robust standard errors in parentheses

The regression analysis reveals a strong positive relationship between GDP and export value, confirming findings from previous studies by Diks and Panchenko (2005) and Kim and Link (2009) Specifically, a one percent increase in the GDP of an exporting country is associated with an approximate 0.623 percent rise in export value Additionally, the export value is influenced not only by the exporting country's GDP but also by the GDP of the importing partner If the GDP of the importing country increases by one percent, the exporting country’s exports are projected to rise by about 0.696 percent This suggests that larger economies tend to produce more goods for export while also generating higher demand for imports, indicating that economic growth fosters bilateral trade between nations.

The relationship between population and export is complex and varies among different estimators While Ordinary Least Squares (OLS) indicate a negative and insignificant correlation for exporting populations, random effect and Hausman-Taylor estimates reveal a positive and significant relationship In contrast, most population variables in importing countries appear insignificant in fixed and random effect estimators, but OLS and Hausman-Taylor show positive and significant results Historical literature and empirical studies, such as those by Thirlwall (1994) and Becker et al (1999), suggest that population can have both negative and positive effects on exports and economic growth A larger population can enhance the labor force and boost domestic consumption, fostering competition and skill development However, it may also lead to challenges such as food scarcity, savings development, and unemployment (Meier, 1995) Moreover, studies by Simon (1992) and Kelley & Schmidt (1996) indicate no statistical correlation between export, economic growth, and population This paper's regression results demonstrate both positive and negative relationships, with the Hausman-Taylor estimator predicting that a 1% increase in the exporting country's population could lead to a 0.43% increase in exports, while a similar increase in the importing country's population could result in a 0.403% rise in exports.

Conclusion and policy recommendation

This study examines the impact of the ASEAN-Japan Comprehensive Economic Partnership (AJCEP) on trade dynamics, specifically focusing on trade creation and diversion effects among ASEAN countries and Japan Utilizing both aggregated and disaggregated datasets for five product categories—agricultural goods, manufactured products, chemical products, machinery and transportation equipment, and textiles—this analysis spans the period from 2000 to 2015, involving nine ASEAN nations and their 14 largest trading partners Employing a gravity model alongside various estimators such as OLS, FEM, REM, and Hausman-Taylor, the research prioritizes the Hausman-Taylor estimator to effectively address country-specific unobserved effects while ensuring consistent estimation for time-invariant variables.

The estimation results indicate that the GDP of both importing and exporting countries, as well as shared borders, positively influence bilateral trade Additionally, cultural factors, including language and colonial ties, also contribute to fostering trade between nations Conversely, geographic factors such as distance and landlocked status serve as significant trade-restrictive elements, exerting a strong negative impact on trade This negative relationship between geographic distance and trade aligns with the hypotheses of the international trade gravity model.

Despite the elimination of tariff commitments, the AJCEP has not effectively promoted trade among member countries or with external partners, as indicated by aggregated total export data Furthermore, disaggregated data reveals that the trade agreement between ASEAN and Japan, known as AJCEP, has a negative or negligible impact on trade across various categories, including agricultural goods, manufactured products, chemical products, machinery and transportation equipment, as well as clothing, accessories, and textiles This negative effect suggests that the removal of tariffs under AJCEP has failed to stimulate intra-bloc trade among AJCEP members and trade flows to external partners.

Japan's involvement in the AJCEP trading agreement has yielded minimal outcomes, indicating that it has not significantly enhanced the integration of AJCEP members into international trade.

Over the past ten years, there has been a significant increase in both bilateral and multilateral free trade agreements, particularly within the ASEAN region, highlighting the crucial role of ASEAN members in global trade However, these agreements have yet to fully realize regional liberalization, as evidenced by the AJCEP The limitations of current FTAs can be attributed to several factors.

Current FTAs lack sufficient liberalization, leading to the need for new agreements that prioritize greater tariff elimination compared to existing ASEAN FTAs + 1.

The complexity of Free Trade Agreements (FTAs) in the ASEAN region arises from an overwhelming number of Rules of Origin (RoO), which vary significantly between agreements, complicating trade management The challenge lies in determining the eligibility for preferential treatment across different products, creating difficulties not only for exporters and manufacturers but also for customs authorities tasked with compliance verification To alleviate administrative burdens and eliminate the "noodle bowl" effect, new FTAs should prioritize simplicity in their rules, ensuring they are transparent, straightforward, and consistent to promote full trade liberalization.

The focus has shifted to trade facilitation and non-tariff trade barriers (NTBs) in recent negotiations of bilateral and multilateral Free Trade Agreements (FTAs) The pace of these negotiations is crucial for developing more appealing trade packages While eliminating tariffs was once the primary concern, discussions now increasingly address NTBs, including food safety standards, technical requirements, customs procedures, and sensitive product lists.

Therefore, ASEAN should pay attention to the new trend of international development and the competitiveness of new regional initiatives to optimize the positive effect the integration causes.

Before ending this paper, we would like to show out the limitation of this study for future research on the impact of AJCEP to international trade.

One significant limitation of the AJCEP is its relatively short duration of implementation, having come into force in December 2008, compared to the ACFTA and AKFTA, which began in 2002 and 2007, respectively This limited timeframe, spanning from 2008 to 2015, restricts our ability to fully assess the long-term impacts of AJCEP on ASEAN and Japan Furthermore, the target elimination period for AJCEP extends to 2018, leaving us without a comprehensive understanding of its overall effects.

One significant limitation of the study is the construction of the database, which lacks the inclusion of critical variables that could substantially influence bilateral trade Notably, this includes control variables for non-trade barriers, administrative costs, rules of origin, and food safety standards.

One significant limitation of the AJCEP (ASEAN-Japan Comprehensive Economic Partnership) is the precision regarding its content Unlike some free trade agreements (FTAs) that implement immediate tariff eliminations upon enforcement, AJCEP features a commitment schedule for the gradual removal of tariffs To fully understand the impact of AJCEP and similar FTAs, it is essential to incorporate these phased schedules into the analysis of their effects on tariff elimination over time.

Aiken, D E 1973 Proecdysis, setal development, and molt prediction in the American lobster (Homarus americanus) J Fish Res Board Can 30: 1337–1344.

Anderson, J E., & Marcouiller, D (2002) Insecurity and the pattern of trade: An empirical investigation Review of Economics and statistics, 84(2), 342-352.

Anderson, J.E and E Van Wincoop, 2003, Gravity with Gravitas: A Solution To The Border Puzzle, American Economic Review 93, pp 170–192.

Antonucci, D., & Manzocchi, S (2006) Does Turkey have a special trade relation with the EU?:

A gravity model approach Economic Systems, 30(2), 157-169.

Baier, S.L and J.H Bergstrand, 2009, Estimating the Effects of Free Trade Agreements on International Trade

Baier, Scott L & Jeffrey H Bergstrand (2004) ‘Economic determinants of free trade agreements, Journal of International Economics 64 (1), 29–63 (October).

Baldwin, R., & Taglioni, D (2006) Gravity for dummies and dummies for gravity equations (No w12516) National Bureau of Economic Research.

Becker, G S., Glaeser, E L., & Murphy, K M (1999) Population and economic growth The American Economic Review, 89(2), 145-149.

Ben-Basaṭ, A (2002) The Israeli Economy, 1985-1998: From Government Intervention to Market Economics MIT Press.

Bergstrand, J H (1985) The gravity equation in international trade: some microeconomic foundations and empirical evidence The review of economics and statistics, 474-481.

Brakman, S., Garretsen, H., & Van Marrewijk, C (2001) An introduction to geographical economics: Trade, location and growth Cambridge university press.

Carrere, C (2006) Revisiting the effects of regional trade agreements on trade flows with proper specification of the gravity model European Economic Review, 50(2), 223-247.

Deardorff, A (1998) Determinants of bilateral trade: does gravity work in a neoclassical world?.

In The regionalization of the world economy (pp 7-32) University of Chicago Press.

Diks, C., & Panchenko, V (2005) A note on the Hiemstra-Jones test for Granger non-causality. Studies in Nonlinear Dynamics & Econometrics, 9(2).

Disdier, A C., & Mayer, T (2007) Je t'aime, moi non plus: Bilateral opinions and international trade European Journal of Political Economy, 23(4), 1140-1159.

Eaton, J., & Kortum, S (2002) Technology, geography, and trade Econometrica, 70(5), 1741- 1779.

Egger, Hartmut, Egger, Peter, Greenaway, David, 2008 The trade structure effects of endogenous regional trade agreements Journal of International Economics 74, 278–298

Egger, P., (2005) ‘Alternative techniques for estimation of cross-section gravity models,’ Review of International Economics, 13(5), 881-891.

Eichengreen, B., & Irwin, D A (1998) The role of history in bilateral trade flows In The regionalization of the world economy (pp 33-62) University of Chicago Press.

Elliott, R J., & Ikemoto, K (2004) AFTA and the Asian Crisis: Help or Hindrance to ASEAN Intra-Regional Trade? Asian Economic Journal, 18(1), 1-23.

Flowerdew, R., & Aitkin, M (1982) A method of fitting the gravity model based on the Poisson distribution Journal of regional science, 22(2), 191-202.

Flows Using Matching Econometrics, Journal of International Economics 77, pp 63–76.

Frankel, J., Stein, E., & Wei, S J (1995) Trading blocs and the Americas: The natural, the unnatural, and the super-natural Journal of development economics, 47(1), 61-95.

Frankel, Jeffrey A., 1997 Regional Trading Blocs Institute for International Economics, Washington, DC.

Fukunaga, Y., & Isono, I (2013) Taking ASEAN+ 1 FTAs towards the RCEP: a mapping study. ERIA Discussion Paper, 2.

Gómez-Herrera, E (2013) Comparing alternative methods to estimate gravity models of bilateral trade Empirical Economics, 44(3), 1087-1111.

Hausman, A & E Taylor (1981) ‘Panel data and unobservable individual effects,’ Econometrica

Head, K., & Mayer, T (2004) The empirics of agglomeration and trade Handbook of regional and urban economics, 4, 2609-2669.

Heckman, J J (1979) Statistical models for discrete panel data Department of Economics and Graduate School of Business, University of Chicago.

Heckman, James, Lochner, Lance, Taber, Christopher, 1998 General-equilibrium treatment effects: a study of tuition policy American Economic Review 88 (2), 381–386 May.

Helpman, E., & Krugman, P R (1985) Market structure and foreign trade: Increasing returns, imperfect competition, and the international economy MIT press.

Helpman, E., Itskhoki, O., & Redding, S (2008) Wages, unemployment and inequality with heterogeneous firms and workers (No w14122) National Bureau of Economic Research.

Kelley, A C., & Schmidt, R M (1996) Saving, dependency and development Journal of Population Economics, 9(4), 365-386.

Kimura, F & H Lee (2006) ‘How does economic freedom affect international trade in services?

Krugman, P (1980) Scale economies, product differentiation, and the pattern of trade The American Economic Review, 70(5), 950-959.

Ku, H., & Zussman, A (2010) Lingua franca: The role of English in international trade Journal of Economic Behavior & Organization, 75(2), 250-260.

Leamer, E E., & Levinsohn, J (1995) International trade theory: the evidence Handbook of international economics, 3, 1339-1394.

Linnemann, Hans, 1966 An Econometric Study of International Trade Flows NorthHolland, Amsterdam.

MacKellar, L., Wửrgửtter, A., & Wửrz, J (2002) Economic growth of landlocked countries In ệkonomie in Theorie und Praxis (pp 213-226) Springer Berlin Heidelberg.

Manchin, M., & Pelkmans-Balaoing, A O (2007) Rules of origin and the web of East Asian free trade agreements World Bank Policy Research Working Paper, (4273).

Martin, W., & Pham, C S (2008) Estimating the gravity equation when zero trade flows are frequent MPRA Paper.

Mayer, T., & Zignago, S (2011) Notes on CEPII’s distances measures: The GeoDist database

McCombie, J S., Thirlwall, A P., & Thompson, P (1994) Economic growth and the balance-of- payments constraint New York: St Martin's press.

Medalla, E M (2011) Taking stock of the ROOs in the ASEAN+ 1 FTAs: toward deepening East Asian integration (No DP 2011-36) Philippine Institute for Development Studies

Melitz, M J (2003) The impact of trade on intra industry‐ reallocations and aggregate industry productivity Econometrica, 71(6), 1695-1725.

Nellis, J G., & Parker, D (2004) Principles of macroeconomics Pearson education.

Oguledo, V., & MacPhee, C R (1994) Gravity models: a reformulation and an application to discriminatory trade arrangements Applied Economics, 26(2), 107-120.

Okabe, M (2015) Impact of Free Trade Agreements on Trade in East Asia(No DP-2015-01).

Silva, J S., & Tenreyro, S (2011) Further simulation evidence on the performance of the Poisson pseudo-maximum likelihood estimator Economics Letters, 112(2), 220-222.

Soloaga, I and A Winters, 2001, How Has Regionalism in The 1990s Affected Trade?, North American Journal of Economics and Finance 12, pp 1–29.

Soloaga, I., & Wintersb, L A (2001) Regionalism in the nineties: What effect on trade? The North American Journal of Economics and Finance, 12(1), 1-29.

Tinbergen, J (1962) ‘Shaping the World Economy: Suggestions for an International Trade Policy’, New York: Twentieth Century Fund Press.

Urata, S., & Okabe, M (2014) Trade Creation and Diversion Effects of Regional Trade Agreements: A Product le‐ vel Analysis The World Economy, 37(2), 267-289.

Viner, J (1950) The economics of customs unions The customs union issue, 41-81.

APPENDIX: Details Estimation results for selected sample

Table 15: OLS regression result (Total export)

Log total export Coefficient Robust Std.

Log Population exporting country -0.0049469 0.0267363 -0.19 0.853 -0.0573625 0.0474687 Log Population importing country -0.1045923 0.0226067 -4.63 0.000 -0.1489121 -0.0602726

Table 16: Fixed effect model regression estimation result (Total export)

Fixed-effects (within) regression Obs per group: min = 1

Group variable: id avg = 14.5 max = 16

R-sq: within = 0.4742 F(7,324) = 93.90 between = 0.4601 Prob > F = 0.0000 overall = 0.4847

Table 17: Random effect model regression estimation result (Total export)

Random-effects GLS regression Obs per group: min = 1

R-sq: within = 0.4721 F(7,324) = 93.90 between = 0.7068 Prob > F = 0.0000 overall = 0.6977 corr(u_i, X) = 0 (assumed) Wald chi2(12) = 985.13

Log total export Coefficient Robust

Table 18: Hausman-Taylor estimator (Total export)

Number of obs = 4701 Number of groups = 325

Hausman-Taylor estimation Obs per group: min = 1

Log total export Coefficient Robust

Log Population exporting country 0.4296208 0.0912479 4.71 0.000 0.2507782 0.6084634 Log Population importing country 0.4026711 0.0969341 4.15 0.000 0.2126838 0.5926584

Table 19: Hausman-Taylor estimator (Total agricultural goods export)

Log total agricultural goods export Coefficient

RobustStd Err t P>|t| [95% Conf Interval]

Table 20: Hausman-Taylor estimator (Total manufactured products export)

Log total manufactured goods export Coefficient Robust Std.

Table 21: Hausman-Taylor estimator (Total chemical products export)

Log total chemical product export Coefficient Robust

Table 22: Hausman-Taylor estimator (Total machinery and transport equipment products export)

Log total machinery and transport equipment product export

Table 23: Hausman-Taylor estimator (Total Cloth, accessories and textiles and fabric products export)

Prob > chi2 = 0.0000Log total Cloth, accessories and textile and fabric products export

Tiêu đề	The Effect of Regional Trade Agreement to Trade Flow: Evidence of Trade Creation and Trade Diversion of ASEAN – Japan Free Trade Agreement
Tác giả	Pham Thi Hien
Người hướng dẫn	Prof. Dr. Nguyen Trong Hoai
Trường học	University of Economics
Chuyên ngành	Development Economics
Thể loại	Thesis
Năm xuất bản	2016
Thành phố	Ho Chi Minh City

Định dạng
Số trang	85
Dung lượng	286,08 KB