Determines the Impact of Economic growth components on the IIT, HIIT, and VIIT between China and South Asian countries.

doi:10.21203/rs.3.rs-2332250/v1

Bilateral trade has significantly influenced economic growth over the past few decades. This study examines bilateral intra-industry trade between China and South Asian countries from 2000 to 2020. Intra-industry trade IIT is further decomposed into vertical intra-industry VIIT and horizontal intra-industry HIIT to indicate two-way and one-way trade relationships. The estimated results show that in the South Asian region, only two countries, Pakistan and India, have a two-way bilateral trade relationship with China, and the intra-industry trade between China and South Asian countries rapidly increased from 2000 to 2020, which positively impacted bilateral economic growth between China and South Asian countries. The findings of the GLS model examine Bilateral cross-section GDP, GDP per capita, FDI, and Min. GDP and Language have a positive and significant impact on IIT but a negative impact on both HIIT and VIIT but the Bilateral Trade cross-section difference FDI, Max. GDP and LXMGDP have a negative influence on IIT but have a positive and significant influence on both HIIT and VIIT.

IIT

The Intra-industry trade

HIIT

Horizontal intra-industry

VIIT

Vertical intra-industry

Economic growth

Bilateral Export and Import

This study investigates drivers the intra-industry trade (IIT), Horizontal intra-industry trade (HIIT), and Vertical intra-industry trade (VIIT) to assess the determinants of bilateral trade relationships and international trade one-way and two-way dimensions, HIIT can be measured to subtract intra-industry trade (IIT) into total trade (TT), while VIIT we get to subtract HIIT into IIT, the second part of the study we also check the impact of bilateral cross-section economic measurement components (${\text{A}\text{v}\text{g}. \text{F}\text{D}\text{I}}_{\text{a}\text{b}}$, ${\text{D}\text{i}\text{f}\text{f}. \text{F}\text{D}\text{I}}_{\text{a}\text{b}}$, ${\text{M}\text{i}\text{n}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}},$ ${\text{M}\text{a}\text{n}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$, ${\text{D}\text{i}\text{f}\text{f}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$, ${\text{D}\text{i}\text{f}\text{f}. \text{G}\text{D}\text{P}\text{P}\text{C}}_{\text{a}\text{b}}$, ${\text{A}\text{v}\text{g}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$,${\text{A}\text{v}\text{g}. \text{G}\text{D}\text{P}\text{P}\text{C}}_{\text{a}\text{b}}$, ${\text{L}. \text{D}\text{i}\text{s}\text{t}\text{a}\text{n}\text{c}\text{e}}_{\text{a}\text{b}}$, ${\text{X}\text{M}\text{G}\text{D}\text{P}}_{\text{a}\text{b}}$, ${\text{L}\text{a}\text{n}\text{g}\text{u}\text{a}\text{r}\text{e}}_{\text{a}\text{b}}$ and ${\text{D}\text{I}\text{S}\text{T}\text{X}}_{\text{a}\text{b}}$) on IIT, HIIT, and VIIT between China and South Asian countries during the period of 2000 to 2020. (Haddou and Jang 2018) the intra-industry trade IIT examines one country’s economy with the world global economy chain. They calculate in intra-industry trade IIT of Algeria with Saudi Arabia, and Korea with MENA countries, results showed that Korea’s intra-industry trade IIT did not affect much with Algeria and Saudi Arabia, but Korea economy integrated with Europe counties strongly. (Gallucci, Dimitrova et al. 2019) examine the intra-industry trade IIT impact on eco-innovation and socio-economic performance of Europe countries over the period of 2010 to 2018, results showed that eco-innovation has a strong close impact on the intra-industry trade IIT. The economic development of a nation is positively impacted by bilateral commerce (Kang, Kennedy et al. 2009). A country's production can be increased in large part through export (Yusoff and Nuh 2015). For several decades, especially among industrialized nations, it has been noted that the concentration of global trade flows within industries as opposed to between industries is an increasing tendency. Intra-industry trade (IIT) or two-way trade has been used to describe the consequent simultaneous export and import by a country of goods inside a specific industry grouping. The basis for inter-industry trade is the disparity in factor endowments between nations. IIT typically takes place between nations with comparable factor endowments. IIT, or the exchange of like-minded goods, is crucial to the growth of global commerce. (Xie, Huang et al. 2020) Examine that intra-industry trade IIT increase year by year of Guangzhou, and automobile industry, Trade imbalance, and Expenditures have a positive impact on intra-industry trade IIT. By applied a multiple regression model and tested multicollinearity and Heteroskedasticity.

According to (OEC 2022) in August 2022 China has a positive trade balance which has a total worth of 79.4 billion dollars including 315 billion dollars in export which is 9.27% less than in 2021 and 236 billion dollars in imports worth, the top main export products of China in 2022 are Telephones which have a total worth of 17.1 billion dollars, Computers have a total worth of 15.7 billion dollars, Integrated Circuits export worth have 11.9 billion dollars, Semiconductor Devices have total 5.97 billion dollars export worth total exports worth and Electric Batteries have a total worth of 5.64 billion dollars. China invested 187.5 billion dollars in South Asian countries from 2021 to 2022. India on the sixth biggest export partner in 2022 with around 1 billion dollars export worth. The South Asia area, which included Nepal, Bhutan, India, the Maldives, Pakistan, Afghanistan, Bangladesh, and Sri Lanka, was created in 1985. According to data from 2020, there are 1.94 billion people living there. The World Bank estimates that South Asia's overall GDP will reach $4.09 trillion in 2021, an increase of 8.9%. According to reports, India has the largest economy in South Asia with a GDP of 3173.39 billion US dollars, followed by Bangladesh (416.27 billion) and Pakistan (346.34 billion). Additionally, South Asia's GDP per capita reached 2176 US dollars. India is regarded as having the biggest and strongest economy among these eight nations, accounting for 78% of South Asia's total GDP. If we compare the GDP growth rate of India with the combined GDP growth rate of the South Asian region, India reported an 8.9% growth rate as against the South Asian GDP growth rate of 8.3%. This is the highest GDP growth rate after 1988 (9.6%), the second largest economy in South Asia is Bangladesh with a total GDP of 416.27 billion dollars reported a GDP growth of 6.9% which is less than India and aggregate GDP growth of South Asia. However, the per capita GDP was reported at 2503 UD dollars which was more than the aggregate GDP of South Asia. The highest per capita GDP is reported by the Maldives with 8994 US dollars followed by Sri Lanka with 3814 US dollars and then Bhutan with 3000 US dollars. However, Afghanistan reported the lowest GDP per capita of 516 UD dollars. (Clark and Stanley 1999) explain the Bilateral Trade relationship between the United States and South and North developing countries by applied the intra-industry trade IIT approach. Results explain after applied Random effect and Groupwise Heteroskedasticity model intra-industry trade IIT impact positive on the bilateral Trade relationship between the United States and developing countries.

Based on various assumptions regarding product differentiation, new trade theory provides a number of models for explaining IIT. The typical conclusions on the role of factor endowments and scale economies in the context of horizontal product differentiation result from the monopolistic competition paradigm. The factors that influence the intra-industry trade have been divided into two groups in previous theoretical and empirical studies: (Marius-Răzvan and Camelia 2015) determine the Bilateral Trade relationship between Romania and 13 European countries to check the intra-industry trade index GL of motor vehicle parts and accessories sector from 1995 to 2012. They compare the intra-industry trade index IIT from 1995 to 2012, their results explain that in 1995 Romania had a high IIT with Poland, the Czech Republic, and Hungary which changes in 2012 Romania’s intra-industry trade IIT with Belgium, Spain, Italy, Bulgaria, Poland, and Czech Republic is high. They applied the Panel Generalized Method of Moments (GGM) to check the intra-industry trade IIT impact on GDP per capita and development expenditures. Results examine the IIT has a positive and significant impact on the economic growth of Romania. (Menon and Dixon 1996) developed the total trade growth TT, IIT, and net trade NT to explain the Relations Trade Agreement between Australia and New Zealand of manufacturing industries during the period of 1981 to 1991. Estimated results examine that IIT contributed growth of more than 54.79 percent out of 97.12 where RTA contributed 4.55% in growth over a period of 1986 to 1991 of Australian TT, but in New Zealand total IIT contributed 28.58 percent and RTA contributed 16.88%. (Ambroziak 2012) Investigate the foreign direct investment FDI impact on the intra-industry trade IIT, vertical intra-industry VIIT and horizontal intra-industry HIIT over the period of 1995 to 2008 to Poland, Hungary, Slovakia, and the Czech Republic. Estimated results of HIIT and VIIT after applied fixed and random effect models explain that GDP and GDP per capita have a positive and significant impacts on both HIIT and VIIT, but Distance and Foreign Direct investment FDI have a negative impact on HIIT and positive impact on VIIT.

According to (Leitão and Balogh 2020) the relationship between climate change and agriculture intra-industry trade by applied fixed effect and GMM model over the period of 2000 to 2014. They showed a negative and significant impact of international trade on climate change, results also explain that agricultural intra-industry trade IIT has a negative impact on climate change. (Sari, Yuliati et al. 2019) the import of textile intra-industry IIT on the economic growth of Bilateral Indonesia and China during the period of 2000 to 2013. Results explain that IIT has a 53.9 percent impact between China and Indonesia. (Thorpe and Leitão 2013) Investigate the impact of Bilateral intra-industry trade between the United States and 16 trade major trade partners over the period of 1995 to 2008. They divide IIT into two parts the vertical VIIT and horizontal HIIT to test the impact more clearly, they also applied pooled OLS, Random and Fixed effects to check the IIT, HIIT, and VIIT impact on Bilateral economic growth between the United States and 16 major trade partner including China and neighborhood countries. They calculate VIIT to subtract HIIT from IIT. Results examine that the difference in GDP and distance per difference in GDP between the USA and trade partners have a negative and significant impact on both HIIT and IIT but have a positive impact on VIIT. Minimum, average and maximum GDP have a positive impact on all IIT, HIIT and VIIT. (Yuan and Du 2020) the Bilateral Trade relationship between China and Japan & South Korea by applied intra-industry trade IIT over the period of 2009 to 2018 of manufacturing products. Results explain that after China changes its trade structure they have a significant impact on intra-industry trade in the Bilateral Trade relationship between China and Japan & South Korea.

(Ferto and Bojnec 2016) the food intra-industry trade of 27 European union counties during the period of 2000 to 2011. Results explain the both the vertical intra-industry VIIT and horizontal intra-industry HIIT positively impact on the economic growth. They also used the pooled OLS model to check the relationship between the vertical intra-industry trade VIIT and horizontal intra-industry HIIT and on the economic growth, results showed that there have a positive and significant impact of GDP per capita and GDP on both VIIT and HIIT. (Van, Van et al. 2020) the intra-industry trade IIT impact on Vietnam’s economic growth over the period of 2000 to 2014. They used the Fixed and Random effect models to check the intra-industry trade IIT impact on Vietnam’s economic growth, results explain that GDP has a negative and significant impact on the intra-industry trade IIT and foreign direct investment FDI has a positive impact in Fixed effect on the intra-industry trade IIT and negative impact in Random effect on IIT. (Ali, H. Khan et al. 2022) the intra-industry trade IIT, Export diversification, and export growth of long-term and short-term impact on Bangladesh’s economic growth over the period of 1990 to 2017. Results examine that there have a long-run relationship between the intra-industry trade index GL and the economic growth of Bangladesh. (Jing, Leitão et al. 2010) calculate the intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT to check the Bilateral Trade relationship between China and trade partners countries, and impact of the intra-industry trade IIT impact on the economic growth of China, results showed that the difference in GDP between China and trade partners have a negative impact on the intra-industry trade IIT and average in GDP between China and trade partners countries, GDP per capita, Distance have a positive impact on the intra-industry trade IIT. (Chemsripong, Lee et al. 2005) the intra-industry trade IIT has a negative and significant impact on transportation and the economic development between Thailand and APEC trade partners countries over the period of 1980 to 1999. (Yaqub, Mehmood et al. 2015) applied the variance inflation factor, the Heteroskedasticity test, the Breusch-Pagan Lagrange Multiplier test, the Wooldridge test of serial correlation and the Hausman test of Fixed and Random effect models to test the impact of stock result on the economic growth. (Saparamadu and Weerasinghe 2021) the Bilateral Trade relationship between Sri Lanka and South Asia countries which are Pakistan, India and Bangladesh by applied the intra-industry trade IIT impact on the economic growth over the period of 1992 to 2017. Results explain the Gross National Income per capita and tariff rate have a negative and significant impact on the intra-industry trade IIT of Sri Lanka. (Kang 2010) there have a positive and significant import of 31 OECD Central and East European countries of GDP export, import and GDP per capita on the intra-industry trade IIT but GDP per capita of import have a negative impact on the intra-industry trade IIT, and FDI also have a positive impact on the intra-industry trade IIT by applied the Pooled and Fixed effect. (Sonrası, Otomotiv et al. 2019) investigate the intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT of the United States over the period of 2007 to 2017. Results showed that the intra-industry trade decrease from recent years. (Crespo and Fontoura 2004) check that there has a positive intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT impact of Portugal main trade partners on the economic growth in end of 1990. (Jambol and Ismail 2013) the intra-industry trade IIT of Malaysian manufacturing industry, machinery and transport equipment over the period of 1990 to 2010 with ASEAN free trade area trade partners, estimated results showed that the intra-industry trade IIT increase from 0.27 to 0.51 from 1990 to 2007 after that last three years it decreased from 0 51 to 0.46. (Brülhart 2009) this paper provides an in-depth analysis of the intra-industry trade patterns and trends. If measured at the highest statistical aggregation five digits level, 27 percent of global trade in 2006 was the intra-industry trade if measured at a lower statistical aggregation three digits level, 44 percent was. Since the early 1960s, there has been a steady increase in the global intra-industry trade, which suggests that there is a process of global structural convergence wherein economies are becoming more similar over time in terms of their sectoral compositions. This trend seems to have been a significantly influenced by the global fragmentation of vertical production chains, especially since the 1990s. African trade is still predominately of the inter-industry type, with intra-industry trade being a phenomenon of high and middle-income countries. Based on the literature in the study, there is a gap that we are investigating:

To determine whether there is a positive or negative relationship between IIT, HIIT & VIIT and the difference in GDP & GDP per capita of China and South Asian countries.

To investigate whether there is a positive or negative relationship between IIT, HIIT & VIIT and the average in GDP & GDP per capita of China and South Asian countries.

To examine Max. GDP & Min. GDP impact on the intra-industry trade IIT, HIIT & VIIT.

To explain whether Openness up has a positive or negative impact on IIT, HIIT & VIIT.

To explore the average and difference in FDI impact on IIT, HIIT & VIIT.

To scrutinize the Language impact on IIT, HIIT & VIIT.

To disclose whether Distance and DISTX have a positive or negative relationship with IIT, HIIT & VIIT.

Study design

Using a quantitative study strategy, this study was conducted to fulfill the goals of investigating how ${\text{A}\text{v}\text{g}. \text{F}\text{D}\text{I}}_{\text{a}\text{b}}$, ${\text{D}\text{i}\text{f}\text{f}. \text{F}\text{D}\text{I}}_{\text{a}\text{b}}$, ${\text{M}\text{i}\text{n}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}},$ ${\text{M}\text{a}\text{n}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$, ${\text{D}\text{i}\text{f}\text{f}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$ ,${\text{D}\text{i}\text{f}\text{f}. \text{G}\text{D}\text{P}\text{P}\text{C}}_{\text{a}\text{b}}$ ,${\text{A}\text{v}\text{g}. \text{G}\text{D}\text{P}}_{\text{a}\text{b}}$ ,${\text{A}\text{v}\text{g}. \text{G}\text{D}\text{P}\text{P}\text{C}}_{\text{a}\text{b}}$, ${\text{L}. \text{D}\text{i}\text{s}\text{t}\text{a}\text{n}\text{c}\text{e}}_{\text{a}\text{b}}$, ${\text{X}\text{M}\text{G}\text{D}\text{P}}_{\text{a}\text{b}}$, ${\text{L}\text{a}\text{n}\text{g}\text{u}\text{a}\text{r}\text{e}}_{\text{a}\text{b}}$ and ${\text{D}\text{I}\text{S}\text{T}\text{X}}_{\text{a}\text{b}}$ impact on Intra-industry trade IIT, Horizontal on Intra-industry trade HIIT, and Vertical on Intra-industry trade VIIT. For this purpose, time series annual data from 2000 to 2020 between China and South Asian countries (Pakistan, Afghanistan, Bangladesh, Bhutan, India, Sir Lanka, Nepal, and Maldives) has been used to investigate trend analysis. Times series analysis deals with the data points collected at consistent intervals or periods over time. The data was obtained from World Bank and World Integrated Trade Solution for the period of 2000 to 2020 (TheWorldBank 2022) and (WITS 2022).

Theoretical Model

Literature about international and bilateral trade gave us knowledge to measure different methods which we can follow to check the bilateral trade relationship between China and South Asian countries. Based on the literature the intra-industry trade model is the first test used in the study examine by (Grubel and Lloyd 1975), they gave a new concept to measure international trade by using intra-industry trade (IIT). (Krugman 1979), (Lancaster 1980) and (Hamilton and Kniest 1991) explain the existence of intra-industry trade and improvement in the IIT model to measure the international trade relationship between two nations, results examine that higher intra-industry trade values showed that strong two-way trade relationship between two countries. (Greenaway, Hine et al. 1995) Investigate the intra-industry trade IIT and gave a new concept to divide IIT into horizontal intra-industry HIIT, and examine methods to calculate the HIIT and VIIT, and vertical intra-industry VIIT to measure the international trade of the United Kingdom, and check the importance of HIIT and VIIT in U.K. (Ekanayake 2001), (Schmitt and Yu 2002), (Metz 2003) and (Fertő, Podruzsik et al. 2016) applied horizontal HIIT and vertical intra-industry VIIT to measure the bilateral trade between different countries and regions, results examined that both HIIT and VIIT are significantly impacted on nations economic growth, these reasons we are applying intra-industry trade IIT into horizontal HIIT and vertical intra-industry VIIT to investigate the bilateral trade relationship between China and South Asian countries. (Varma 2015) and (Jambor 2014) both HIIT and VIIT are equally impact for a country to increase bilateral trade. (Chang 2009) and (Ferto and Jambor 2015) examine that VIIT has negative effect on economic growth but HIIT has a positive effect.

The study also investigates the economic growth measurement component’s impact on intra-trade, horizontal and vertical intra-industry trade, we used a multiple regression model based on the Gravity model by (Anderson 1979), applied the Gravity model to test border and neighboring countries’ effect on international trade. (Feenstra 2002) and (Okubo 2004) examine the effect of international trade by used the Gravity model on Japan market. Results showed that Canada and the United State have more neighboring countries’ effects than Japan. (Gómez-Herrera 2013), (Egger, Francois et al. 2017), (Jalerajabi and Moghaddasi 2017) and (Yang, Anderson et al. 2020) investigate the bilateral trade impact on economic growth by used a multiple regression model based on the Gravity model. Results mention that international trade has a positive impact on the economic growth of countries. (Clark and Linzer 2015), (Gardiner, Luo et al. 2009), (Doucouliagos and Ulubaşoğlu 2008), and (Bell, Fairbrother et al. 2019) explain how and why we can applied Fixed and Random effects to test the best fit model for the study to check the effect of independent variables on depended variables. (Glejser 1969) gave a new concept to test the error term in the model by applied the heteroskedasticity test. (Schwert and Seguin 1990), (MacKinnon 2013) and (Baum and Lewbel 2019) year by year improve of the heteroskedasticity test to find the heteroskedasticity problem in the research model which is very useful for researchers nowadays. (Koenig 1999) to test the associability in a long-run relationships they developed a statistical approach and called it Auto-correlation. (Coad and Hölzl 2009) and (Cheng, Haworth et al. 2012) also applied and tested the auto-correlation problems in the model. (Magnus 1978) based on the regression model Magnus in 1978 gave a new concept to solve a few problems in the model that simple regression model cant able to solved and called that approach the GLS model. (Li and Stengos 1994) for unknown errors need to use the GLS model rather than the simple Regression model. (Hayes and Cai 2007), (Gitto, Minervini et al. 2016), (Zhang, Kumbhakar et al. 2021) and (Al Shabasy, Abbassi et al. 2022) to solved the heteroskedasticity, homoscedasticity and auto-correlation problem in the model they applied the GLS model.

Empirical Models

According to this concept, international trade in a broad industrial category that comprises two-way trade inside specifically defined industries is referred to as HIIT. Balanced commerce within a broadly specified industry class, which consists of exports and imports from specifically defined industries, is known as VIIT (Kandogan 2003) and (Thorpe and Leitão 2013).

${TT}_{abm}$ = ${NT}_{abm}$ + ${IIT}_{abm}$ _______________ (1)

${TT}_{abm}$ is Total trade of country a to country b for commodity m.

${NT}_{abm}$ is Net trade of commodity m of country a with country b.

${IIT}_{abm}$ is indicate as Intra-industry trade of commodity m of country a to country b.

${TT}_{abm}$ = ${M}_{abm}$ + ${X}_{abm}$ _________________(2)

Where ${M}_{abm}$ is indicates as total imports of country a from country b, ${X}_{abm}$ is indicates as total exports of country a to country b.

${NT}_{abm}$ = |${X}_{abm}$ - ${M}_{abm}$| __________________(3)

Net trade can mention as total imports subtract with total exports

${INT}_{abm}$ = ${TT}_{abm}$ - ${IIT}_{abm}$_______________(4)

${INT}_{abm}$ is indicates the inter-industry trade, ${TT}_{abm}$ is mention as total trade of country a to country b and ${IIT}_{abm}$ indicates as intra-industry trade.

${INT}_{abm}$ = $H{IIT}_{abm}$_______________(5)

Based on literature, these methods we used to calculate IIT, HIIT, and VIIT to find the bilateral trade relationship between China and South Asian countries. we follow the literature to calculate IIT which mention bellow.

${IIT}_{abm} = \frac{{(M}_{abm} + {X}_{abm})- |{X}_{abm} - {M}_{abm}|}{{M}_{abm} + {X}_{abm}}$ _____________(6)

${IIT}_{abm} = \frac{{(M}_{abm} + {X}_{abm})}{{M}_{abm} + {X}_{abm}}$ - $\frac{{|M}_{abm} + {X}_{abm}|}{{M}_{abm} + {X}_{abm}}$ ____________(7)

${IIT}_{abm} = 1$ - $\frac{{|M}_{abm} + {X}_{abm}|}{{M}_{abm} + {X}_{abm}}$ _______________ (8)

${IIT}_{abm}$ = ${VIIT}_{abm}$ + ${HIIT}_{abm}$ _______________(9)

${IIT}_{abm}$ is indicates as Intra-industry trade, ${HIIT}_{abm}$ is indicates as Horizontal intra-industry trade, and ${VIIT}_{abm}$ is vertical Intra-industry trade.

Based on the Gravity model by (Anderson 1979) in the study we used bilateral trade a multiple regression models of cross-section average, difference and both cross-section average and difference at Fixed & Random effect, and the GLS model (Li and Stengos 1994), (Zhang, Kumbhakar et al. 2021) and (Al Shabasy, Abbassi et al. 2022).

${IIT}_{abm}$ = ${\beta }_{0}$ + ${\beta }_{1}{Avg. FDI}_{ab}$ + ${\beta }_{2}{Min. GDP}_{ab}+$ ${\beta }_{3}{Man. GDP}_{ab}$ + ${\beta }_{4}{Avg. GDP}_{ab}$ + ${\beta }_{5}{Avg. GDPPC}_{ab}$ + ${\beta }_{6}{L. Distance}_{ab}$ + ${\beta }_{7}{XMGDP}_{ab}$ + DISTX + ${\beta }_{8}{Languare}_{ab}$ + ${\beta }_{9}{DISTX}_{ab}$ ___________(10)

${IIT}_{abm}$ = ${\beta }_{0}$ + ${\beta }_{1}{Diff. FDI}_{ab}$ + ${\beta }_{2}{Min. GDP}_{ab}+$ ${\beta }_{3}{Man. GDP}_{ab}$ + ${\beta }_{4}{Diff. GDP}_{ab}$ + ${\beta }_{5}{Diff. GDPPC}_{ab}$ + ${\beta }_{6}{L. Distance}_{ab}$ + ${\beta }_{7}{XMGDP}_{ab}$ + ${\beta }_{8}{Languare}_{ab}$ + ${\beta }_{9}{DISTX}_{ab}$ ___________(11)

${IIT}_{abm}$ = ${\beta }_{0}$ + ${\beta }_{1}{Avg. FDI}_{ab}$ + ${\beta }_{2}{Diff. FDI}_{ab}$ + ${\beta }_{3}{Min. GDP}_{ab}+$ ${\beta }_{4}{Man. GDP}_{ab}$ + ${\beta }_{5}{Diff. GDP}_{ab}$ + ${\beta }_{6}{Diff. GDPPC}_{ab}$ + ${\beta }_{7}{Avg. GDP}_{ab}$ + ${\beta }_{8}{Avg. GDPPC}_{ab}$ + ${\beta }_{9}{L. Distance}_{ab}$ + ${\beta }_{10}{XMGDP}_{ab}$ + ${\beta }_{11}{Languare}_{ab}$ + ${\beta }_{12}{DISTX}_{ab}$ ___________(12)

Where ${Avg. FDI}_{ab}$ indicates as log average foreign direct investment between China and South Asian countries, ${Diff. FDI}_{ab}$ designates as log difference in foreign direct investment between High FDI(China) into low FDI(South Asian countries), ${Min. GDP}_{ab}$ examine log minimum gross domestic products between China and South Asia countries, ${Man. GDP}_{ab}$is a log maximum gross domestic products between China and South Asia countries, ${Diff. GDP}_{ab}$indicates log difference in gross domestic products between high GDP(China) and low GDP(South Asian countries), ${Diff. GDPPC}_{ab}$ designates log difference in gross domestic products per capita between high GDP(China) and low GDP(South Asian countries), ${Avg. GDP}_{ab}$ specifies log average gross domestic products between China and South Asia countries, ${Avg. GDPPC}_{ab}$ indicates log average gross domestic products per capita between China and South Asia countries, ${L. Distance}_{ab}$ indicates log value of distance between China and each South Asian county, ${XMGDP}_{ab}$(Openness) designates sum of log bilateral export and import between China and South Asian countries divided by the bilateral gross domestic product between China and South Asian countries, and ${Languare}_{ab}$ common language between China and South Asian countries, ${DISTX}_{ab}$ indicates as distance between China and South Asian countries multiply by bilateral gross domestic product between China and South Asian countries. ${\beta }_{0}$, ${\beta }_{1}$, ….. ${\beta }_{12}$ indicates as coefficient

Figure 1 explain the bilateral export, import, percentage change of export in the GDP and annul growth of bilateral export and import from 2000 to 2020. Figure 1 showed that the exports of China to South Asia countries increased but import remains unchanged, annual growth of bilateral export and import between China and South Asian countries decreased from 2000 to 2020, but bilateral export parentage change in the GDP increased from 2000 to 2020.

Figure 2 examine the average bilateral intra-industry trade IIT, vertical intra-industry VIIT and horizontal intra-industry HIIT between China and South Asian countries over the period of 2000 to 2020. Results showed that the intra-industry trade IIT and HIIT are decreased but VIIT is increased from 2000 to 2020.

China IIT, HIIT, and VIIT with South Asia trade partners are mentioned in table 1, between 2000 to 2022. According to the formula of the intra-industry trade (IIT), Horizontal Intra-industry trade (HIIT), and Vertical Intra-industry trade (VIIT) level between China and South Asian eight countries is measured in table 1. Estimated results showed how strong is the Bilateral trade relationship between China and South Asian countries.

Table: 1 IIT, HIIT and VIIT of China by South Asian Countries.

	2000			2010			2020
Countries	IIT	HIIT	VIIT	IIT	HIIT	VIIT	IIT		HIIT	VIIT
Afghanistan	0.1964	0.8036	-0.6073	0.0411	0.9589	-0.9177		0.4268	0.5732	-0.1465
Bangladesh	0.1008	0.8992	-0.7985	0.0762	0.9238	-0.8476		0.0410	0.9590	-0.9179
Bhutan	0.0048	0.9952	-0.9904	0.0160	0.9840	-0.9681		0.0191	0.9809	-0.9617
India	0.4784	0.5216	-0.0432	0.6751	0.3249	-0.3501		0.9289	0.0711	0.8578
Maldives	0.0410	0.9590	-0.9180	0.0015	0.9985	-0.9970		0.0299	0.9701	-0.9401
Nepal	0.0273	0.9727	-0.9454	0.0307	0.9693	-0.9385		0.0699	0.9301	-0.8602
Pakistan	0.2431	0.7569	-0.5138	0.3994	0.6006	-0.2013		0.8468	0.1532	0.6935
Sri Lanka	0.1528	0.8472	-0.6945	0.0975	0.9025	-0.8049		0.0561	0.9439	-0.8879

Table 1 explains the Intra-industry trade (IIT), Horizontal Intra-industry trade (HIIT), and Vertical Intra-industry trade (VIIT) values of South Asian countries over the period of 2000 to 2020, highest IIT in 2000 between China and South Asian belongs to India that value is 0.4784 and the lowest belongs to Bhutan which is 0.0048, that rapidly change from to 2000 to 202. Estimated results also showed that their bilateral trade relationship become stronger in recent years during the opening up and Belt and Road initiative. The data give some indications of the kind of structural changes that happened in the bilateral industrial trade of the nation. The external trading environment went through a period of tremendous transformation that had an impact on both China and the world trading environment as a whole (Greenaway, Hine et al. 1995). Given its considerable participation in global commerce, the IIT of China is a key topic for analysis. The fact that separate drivers are believed to be driving HIIT and VIIT has consequences for the adjustment costs inside a country linked with an increase in trade, as has been noted elsewhere (Varma 2015) and (Jambor 2014).

Table: 2 Descriptive Statistic of All Variables.

Variable	Obs	Mean	Std. Dev.	Min	Max
Avg. GDP	168	12.4269	0.3644	11.7824	12.9384
Diff. GDPPC	168	3.3659	0.4487	1.9527	3.9953
Max. GDP	168	12.7072	0.3717	12.0833	13.1670
Min. GDP	166	10.4818	0.9861	8.6278	12.4520
Diff. GDP	168	12.6846	0.3810	11.8710	13.1669
Avg. FDI	168	10.8679	0.2688	10.3232	11.2029
DISTX	168	16.1344	0.4117	15.2862	16.8584
IIT	168	0.1785	0.2416	0.0003	0.9555
HIIT	168	0.8215	0.2416	0.0445	0.9997
VIIT	168	0.7585	0.2655	0.0117	0.9993
Avg. GDPPC	168	3.4332	0.3334	2.7741	4.0151
Diff. FDI	168	11.0359	0.2695	10.4993	11.3526
L. Distance	168	3.4489	0.1577	3.2030	3.6915
Language	168	-6.8997	1.5023	-10.8117	-4.1403

Table 2 showed the descriptive statistic of all variables to measure the total number of observations, Average, Maximum, Minimum, and Standard Deviation to check the nature of the variables.

Table: 3 Panel Unit Root test.

In this part of the study table 3 examine the empirical results of the Panel Unit Root test to check the Stationarity of variables at a level and first difference. For stationarity first-generation we applied the IPS and Fisher-ADF test, and for the second-generation unit root test includes the CIPS and CAFD.

Variables	IPS		FADF		CIPS		CAFD
	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)
Avg. GDP	4.472	-4.211*	5.559	78.413*	-2.013	-3.685*	-5.887*	-5.756*
Diff. GDPPC	4.492	-2.344*	1.971	37.818*	-4.047*	-2.509*	-3.242*	-2.580**
Max GDP	5.532	-1.677**	0.207	23.576***	2.610*	2.610*	2.610	2.610
Min GDP	4.531	-3.181*	2.482	44.964*	-2.349*	-4.443*	-1.903	-2.355**
Diff. GDP	5.498	-1.563*	0.212	22.612	-2.215	-5.088*	-1.463	-3.393*
Avg. FDI	2.749	-4.360*	1.969	55.777*	-2.823*	-4.026*	-5.567*	-5.689*
DISTX	5.498	-1.563*	0.212	22.612	-2.215	-5.088*	-1.462	-3.393*
IIT	-0.754	-3.132*	22.315**	47.817*	-1.531	-4.160*	-1.643	-2.562**
HIIT	-0.754	-3.131*	25.315**	47.816*	-1.531	-4.160*	-1.643	-2.562**
VIIT	-0.754	-3.131*	25.315**	47.816*	-1.531	-4.160*	-1.643	-2.562**
Avg. GDPPC	5.452	-3.995*	2.503	68.543	-2.480**	-5.104*	-3.318*	-4.054*
Diff. FDI	2.922	-4.364*	1.616	55.762*	-2.858*	-4.030*	-5.580*	-5.719*
L.XMGDP	-1.616**	-4.083*	28.728**	64.614*	-1.843	-4.041*	-1.668	-2.683*

Note: * , ** and *** represent 1%, 5% and 10% significance level, respectively

Table 3 represents variables of the research reputed row and Panel unit root test displays in columns. Estimated results of level I(0) and the first-difference I(1) are given for each variable. Results showed that some variables are stationary at the level I(0), while others are stationary at the first difference I(1). In the PLS test of first-generation, all variables are stationary at first difference only LXMGDP is stationary at level. In Fisher-ADF test IIT, HIIT, VIIT and LXMGDP are stationary at the level I(0), and other variables are stationary at the first difference. For second-generation test of CIPS test Diff. FDI, Avg. GDPPC, Diff. FDI, Min. GDP, Max. GDP and Diff. GDPPC are stationary at level but other all variables are stationary at the first difference. The CAFD results of second-generation showed that Avg. GDP, Avg. FDI, Diff. GDPPC, Avg. GDPPC and Diff. FDI are stationary at the level I(0), and other all variables are stationary at the first difference I(1).

Table: 4 Pooled OLS of IIT, HIIT and VIIT of Cross-section Difference.

Variables	IIT	HIIT	VIIT
Min. GDP	0.4582***	-0.4582***	-0.9163***
Max. GDP	-0.9923*	0.9923*	1.9846*
Diff. GDPPC	0.1620***	-0.1620***	-0.3240***
Diff. GDP	0.4380	-0.4380	-0.8761
DISTX	-0.1981	0.1981	0.3963
Diff. FDI	0.0756	-0.0705	-0.1412
L. Distance	0.4265	-0.4265	-0.8530
X+M/GDP	-0.1554***	0.1553***	0.3107***
Language	0.2454***	-0.2454***	-0.4909***
Constant	1.7011*	-0.7012	-2.4023
r-square	0.6922	0.6922	0.6922
Obs.	166	166	166
Pesaran’s cross-sectional	Pr = 0.3224 p-value = 0.377	Pr = 0.3224 p-value = 0.377	Pr = 0.3224 p-value = 0.377
Breusch and Pagan Lagrangian	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

Table 4 explains the Pooled OLS of IIT, HIIT, and VIIT to test the bilateral trade relationship between China and South Asia. Estimated results clarify that the Min. GDP, Diff. GDPPC, Language, and Constant have a significant and positive impact on intra-industry trade IIT but has a negative impact on both HIIT and VIIT at 1% confidence interval. Max. GDP and LXMGDP have a significant and negative impact on IIT and has a positive impact on both HIIT and VIIT. Pasaran’s test result showed that there does not have cross-sectional problem and the Breusch and Pagan test p-value is 1.00 which examine that the Pooled OLX model is best for the study.

Table: 5 Pooled OLS of IIT, HIIT and VIIT of Cross-section Average.

Variables	IIT	HIIT	VIIT
Min. GDP	0.1099*	-0.1099*	-0.2198*
Max. GDP	-56.2231***	56.2231***	112.4462***
Avg. GDPPC	-0.7878***	0.7878***	1.5756***
Avg. GDP	34.4367***	-34.4367***	-68.8734***
DISTX	22.5432***	-22.5433***	-45.0866***
Avg. FDI	-0.1093	0.1093	0.2186
L. Distance	-22.1450***	22.1450***	44.2899***
X+M/GDP	-0.0036	0.0035	0.0071
Language	0.2753***	-0.2753***	-0.5506***
Constant	2.1005	-1.1005	-3.2011
r-square	0.7705	0.7705	7705
Obs.	166	166	166
Pesaran’s cross-sectional	Pr = 0.2905 p-value = 0.299	Pr = 0.2905 p-value = 0.299	Pr = 0.2905 p-value = 0.299
Breusch and Pagan Lagrangian	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

Results of table 5 examines the IIT, HIIT, and VIIT of pooled OLX, estimated results showed that the Min. GDP, Avg. GDP, DISTX and Language have a positive but significant impact on IIT and have a significant and negative impact on HIIT and VIIT. Max. GDP, Avg. GDPPC, Avg. FDI, L. Distance and Language have a negative and significant impact on IIT but have a positive and significant impact on HIIT and VIIT. Pagan test examines that Pooled OLX model is best for the study and there is no cross-section effect in model.

Table: 5 Pooled OLS of IIT, HIIT and VIIT of Cross-section Average and Difference.

Variables	IIT	HIIT	VIIT
DISTX	18.4452***	-18.3418***	-36.6836***
Avg. FDI	0.7138	-0.7138	-1.4276
Diff. GDP	0.5432	-0.5432	-1.2770
Min. GDP	0.1723***	-0.1723***	-0.3446***
Max. GDP	-46.8490***	46.8490***	93.6757***
Diff. GDPPC	0.1214***	-0.1227***	-0.2453***
Avg. GDP	28.5121***	-28.5121***	-57.0242***
Avg. GDPPC	-0.9005***	0.9005***	1.8127***
Diff. FDI	-0.8304*	0.8287*	1.6575*
L. Distance	-17.9433***	17.9403***	35.6631***
X+M/GDP	-0.0239	0.0231	0.04771
Language	0.3471***	-0.3488***	-0.6942***
Constant	0.6975	0.3025	-0.2790
r-square	0.7858	0.7858	0.7864
Obs.	166	166	166
Pesaran’s cross-sectional	Pr = 0.1419 p-value = 0.279	Pr = 0.1419 p-value = 0.279	Pr = 0.1419 p-value = 0.279
Breusch and Pagan Lagrangian	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00	Chibar2(01) = 0.00 Prob > chibar2= 1.00

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

In table 5, we mention both the Average and Difference variables’ impact on IIT, HIIT and VITT. Estimated results explain that the Breusch and Pagan value is greater than 0.05 which examines that this model good fit for the study, Pesaran’s test showed that there does not have cross-section effect. Results also explain that the Min. GDP, Avg. GDP, Diff. GDPPC DISTX and Language have a positively significant impact on IIT but have a negative and significant impacts on both HIIT and VIIT. L. Distance, Avg. FDI, Avg. GDPPC and Max. GDP have a significantly negative impact on IIT but have a positive and significant impact HIIT and VIIT.

Table: 6 Cross-section Average Variables test

Tests Avg.	Fixed Effect (IIT)	Random Effect (IIT)
Pesaran’s cross-sectional Test	Pr = 0.6094 p-value = 0.193	Pr = 0.5973 p-value = 0.400
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.985 p-value = 0.3541
Tests Avg.	Fixed Effect (HIIT)	Random Effect (HIIT)
Pesaran’s cross-sectional Test	Pr = 0.6094 p-value = 0.193	Pr = 0.5973 p-value = 0.400
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.985 p-value = 0.3541
Tests Avg.	Fixed Effect (VIIT)	Random Effect (VIIT)
Pesaran’s cross-sectional Test	Pr = 0.6094 p-value = 0.193	Pr = 0.5973 p-value = 0.400
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.985 p-value = 0.3541

In Table 6, we investigate the average values test, the estimated results examine that the value of the Hausman test in the IIT, HIIT, and VIIT showed that the Fixed effect good fit between Random and Fixed effect, Pesaran test results showed that there is no cross-section in the Fixed and Random effect in the IIT, HIIT and VIIT models, results of the Wald test explain that there had Heteroskedasticity but not have autocorrelation in the motels.

Table: 7 FGLS model of Cross-section Average variables.

Variables	Avg. IIT Cross-sectional t-s FGLS Reg.	Avg. HIIT Cross-sectional t-s FGLS Reg.	Avg. VIIT Cross-sectional t-s FGLS Reg.
DISTX	-0.0732	0.0732	0.1464
Avg. GDP	1.4843	-1.4842	-2.9686
Min. GDP	0.3492***	-0.3492***	-0.6985***
Max. GDP	-1.9679	1.9679	3.9358
Avg. FDI	-0.0585	-0.1465*	-0.2930*
Avg. GDPPC	0.1465*	0.0585	0.1171
L. Distance	0.3798	-0.3798	-0.7596
X+M/GDP	-0.1456***	0.1456***	0.2912***
Language	0.1200***	-0.1200***	-0.2400***
Constant	0.4917	0.5083	0.0165
p-value	0.00	0.00	0.00
Obs.	166	166	166

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

Table 7 example of the GLS model to solve the problem of Heteroskedasticity in average values of the IIT, HIIT, and VIIT. The estimated results explain that Min. GDP and Language have a positive and significant impacts on IIT but have a negative impact on both HIIT and VIIT. Avg. FDI have a significant and negative impact on both HIIT and VIIT, Avg. GDPPC have a positive and significant impact on IIT, LXMGDP has a significantly negative impact on IIT but had a positive impact on both HIIT and VIIT. Total 166 observation we have to test the model.

Table: 8 Cross-section Difference Variables test

Tests Diff.	Fixed Effect (IIT)	Random Effect (IIT)
Pesaran’s cross-sectional Test	Pr = 0.4920 p-value = 0.209	Pr = 0.8025 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.948 p-value = 0.3626
Tests Diff.	Fixed Effect (HIIT)	Random Effect (HIIT)
Pesaran’s cross-sectional Test	Pr = 0.4920 p-value = 0.209	Pr = 0.8025 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.948 p-value = 0.3626
Tests Diff.	Fixed Effect (VIIT)	Random Effect (VIIT)
Pesaran’s cross-sectional Test	Pr = 0.4920 p-value = 0.209	Pr = 0.8025 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.948 p-value = 0.3626

In Table 8, different values test, the estimated results examine that the value of Hausman test in the IIT, HIIT, and VIIT showed that the Fixed effect good for between Random and Fixed effect, Pesaran test results showed that there are no cross-section in the Fixed and Random effect the IIT, HIIT, and VIIT models, results of the Wald test explain that there had Heteroskedasticity but not have autocorrelation in both Fixed and Random effect models.

Table: 9 FGLS model of Cross-section Difference variables.

Variables	Diff. IIT Cross-sectional t-s FGLS Reg.	Diff. HIIT Cross-sectional t-s FGLS Reg.	Diff. VIIT Cross-sectional t-s FGLS Reg.
DISTX	-0.7732	0.7732	1.5464
Diff. GDP	0.4189	-0.4189	-0.8377
Min. GDP	0.4482***	-0.4482***	-0.8963***
Max. GDP	-0.5392	0.5392	1.0783
Diff. FDI	0.1663*	-0.1663*	-0.3325*
Diff. GDPPC	0.1033***	-0.1033***	-0.2067***
L. Distance	1.1574**	-1.1574**	-2.3149**
X+M/GDP	-0.1890***	0.1890***	0.3781***
Language	0.1493***	-0.1493***	-0.2986***
Constant	1.9377**	-0.9377	-2.8753*
p-value	0.00	0.00	0.00
Obs.	166	166	166

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

The results of GLS examine in Table 9 showed that Min. GDP, Diff. GDPPC, Diff. FDI, L. Distance, Constant, and Language have the positive and significant impacts on the IIT but had a negative impact on both HIIT and VIIT. LXMGDP had a negative and significant impacts on IIT but had a positive and significant impact on both HIIT and VIIT of 166 observation.

Table: 10 Both Cross-section Average and Difference Variables test

Tests All	Fixed Effect (IIT)	Random Effect (IIT)
Pesaran’s cross-sectional Test	Pr = 0.6420 p-value = 0.194	Pr = 0.6424 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.978 p-value = 0.3557
Tests All	Fixed Effect (HIIT)	Random Effect (HIIT)
Pesaran’s cross-sectional Test	Pr = 0.6420 p-value = 0.194	Pr = 0.6424 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.978 p-value = 0.3557
Tests All	Fixed Effect (VIIT)	Random Effect (VIIT)
Pesaran’s cross-sectional Test	Pr = 0.6420 p-value = 0.194	Pr = 0.6424 p-value = 0.329
Hausman Test Statistic	p-value = 0.00
Heteroskedasticity Modified Wald Test	p-value = 0.00
Wooldridge Auto-correlation Test	F(1 , 7) = 0.978 p-value = 0.3557

In Table 10, we mention both average and difference values test, the estimated results examines that the value of the Hausman test in the IIT, HIIT, and VIIT showed that the Fixed effect good between Random and Fixed effect, Pesaran test results showed that there are no cross-section in the Fixed and Random effect the IIT, HIIT, and VIIT models, results of the Wald test explains that there had Heteroskedasticity but no autocorrelation in the motels.

Table: 11 FGLS model of both Cross-section Average and Difference variables.

Variables	All IIT Cross-sectional t-s FGLS Reg.	All HIIT Cross-sectional t-s FGLS Reg.	All VIIT Cross-sectional t-s FGLS Reg.
DISTX	0.1007	-0.1007	-0.2014
Avg. FDI	2.3104***	-2.3104***	-4.6207***
Diff. GDP	1.2602*	-1.2602*	-2.5204*
Min. GDP	0.3909***	-0.3909***	-0.7818***
Max. GDP	-4.2223*	4.2223*	8.4445*
Diff. GDPPC	0.1109***	-0.1109***	-0.2217***
Avg. GDP	2.1637	-2.1637	-4.3273
Avg. GDPPC	-0.0575	0.0575	0.1150
Diff. FDI	-2.1778***	2.1778***	4.3556***
L. Distance	0.3391	-0.3391	-0.6781
X+M/GDP	-0.1551***	0.1551***	0.3102***
Language	0.1395***	-0.1395***	-0.2791***
Constant	1.6820	-0.6820	-2.3640
p-value	0.00	0.00	0.00
Obs.	166	166	166

Note: * , ** and *** represent 10%, 5% and 1% significance level, respectively

Table 11 examines the GLS model to solve the problem of Heteroskedasticity in average values of the IIT, HIIT, and VIIT. The estimated results showed that Min. GDP, Avg. FDI, Diff. GDPPC and Language have a positive and significant impacts on IIT but have a negative impact on both HIIT and VIIT. Diff. FDI, Max. GDP and LXMGDP(opening) have a significant and negative impacts on both HIIT and VIIT. Total 166 observation we have to test the model.

The Amis of the study are to investigate the impact of Bilateral intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT impact on Bilateral economic growth between China and South Asian countries (Pakistan, Afghanistan, Bangladesh, Bhutan, India, Sri Lanka, Nepal, and Maldives) over the period of 2000 to 2020. Estimated results showed that the of values intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT increased year by year which examines that the trade relationship between China and South Asian countries rapidly increased from 2000 to 2020, results compared with (Hamilton and Kniest 1991), (Metz 2003) and (Fertő, Podruzsik et al. 2016). Estimated results of first-generation (PLS & FADF) and second-generation (CIPS & CAFD) investigate that all variables are stationary at the first difference I(0) in both first-generation and second-generation tests similar to methods reported in the literature (MacKinnon 2013) and (Baum and Lewbel 2019).

To test the impact of GDP per capita and GDP on intra-industry trade between China and South Asian countries we applied multiple time series Pooled OLS model based on the Gravity model. The cross-section difference methods test between China and South Asian countries examine that the cross-section difference in GDP per capita, Max. GDP, Min. GDP, and Language have a positive and significant impacts on IIT but have a negative impact on both horizontal intra-industry trade HIIT and vertical intra-industry trade VIIT, which explain that intra-industry trade IIT influence Bilateral Trade economic between China and South Asian countries positively but have a negative influence on both HIIT and VIIT in cross-section difference methods test. But dummy variable LXMGDP has a positive impact on intra-industry trade IIT but had a negative impacts on both HIIT and VIIT when talking about the cross-section average method test results showed that Avg. GDP, Min. GDP DISXT and Language have a positive and significant impacts on IIT but a affect negatively on both HIIT and VIIT, and Avg. GDPPC, Max. GDP and L. Distance have a negative impact on intra-industry trade but effect positive on both HIIT and VIIT. Estimated results of both cross-section average and difference methods together to check the Bilateral Trade impact on economic growth than estimated results showed different results which indicate that the DISXT, Avg. GDP, Min. GDP, Diff GDPPC, and Language have a positive and significant impact on IIT but had a negative impact on both HIIT and VIIT, but L. Distance, Avg. FDI, Avg. GDPPC and Max. GDP have a negative impact on intra-industry trade but effect positively on both HIIT and VIIT, which showed that economic indicators have a positive and significant impact on intra-industry trade but effect negatively on both HIIT and VIIT models which are inter-industry trade, but dummy variables affect negatively on IIT but had a positive impact on HIIT and VIIT. The Pesarans text value is not significant which showed that data is not cross-sectional in all three models and the Breusch-Pagan test also has not significant which examines that pooled OLS best fit for the study but there have Heteroskedasticity problems in models for that resin we applied both Fixed and Random effect models similar as (Egger, Francois et al. 2017), (Jalerajabi and Moghaddasi 2017) and (Yang, Anderson et al. 2020)

Results of the Hausman test examine the Fixed effect model is best fit in the study between both Fixed and Random effect models but the Wald test explains that there had a Heteroskedasticity problem in the fixed effect model and the Wooldridge model’s result showed that there not have autocorrelation and cross-section too in the fixed effect. To solve the Heteroskedasticity problem we applied GLS model to test the bilateral economic growth components impact on the intra-industry trade IIT, vertical intra-industry VIIT, and horizontal intra-industry HIIT. Results are based on (Cheng, Haworth et al. 2012), (MacKinnon 2013) and (Baum and Lewbel 2019)

The GLS model in literature researchers used to solve the Heteroskedasticity problem in the models, the estimated results of the cross-section average GLS model investigate that Min. GDP, Avg. GDPPC, and Language have a positive and significant impacts on the intra-industry trade IIT but have a negative impact on both HIIT and VIIT models, and LXMGDP has a negative impact on the intra-industry trade IIT but had a positive impact on both HIIT and VIIT models. The estimated results of the cross-section difference method GLS explain that the Min. GDP, Diff. GDPPC, Diff. FDI, Diff. GDP, Language, and L. Distance have a positively significant impact on IIT but had a negative impact on both HIIT and VIIT models, and LXMGDP has a negative impact on intra-industry trade IIT but had a negative impact on both HIIT and VIIT models. The estimated results of cross-section Avg. FDI Diff. GDPPC, Min. GDP, and Language have a positive and significant impact on IIT but have a negative impacts on both HIIT and VIIT in the GLS model, the LXMGDP, Diff. FDI and Max. GDP have a negative impacts on intra-industry trade IIT and a negatively significant at a 1% confidence interval on HIIT and VIIT, we follow and match with (Gitto, Minervini et al. 2016), (Zhang, Kumbhakar et al. 2021) and (Al Shabasy, Abbassi et al. 2022).

There are some drawbacks to the study. For example, the data sample we used to examine the bilateral trade relations between China and South Asian nations covered the years 2000 to 2020, this informs us that there is a gap in the data sample, which will be filled in the future by using data from different countries and regions, as well as a period that is open to change. In the study, we used the multiple regression OLS and GLS models based on the Gravity model, but the exchange rate effect needs to check on IIT, HIIT, and VIIT in the future, also can use other models and industries products to check the bilateral trade relationship, for example, the Trade Integration Index (TCD), Trade Complementarity Index (TCI), Term of trade, trade intensity index TII, Intra-industry trade index GL and Gravity model, Co-integration test etc. to check the bilateral trade impact on economic growth.

In this study, we investigate the bilateral trade relationship between China and South Asian countries by applied intra-industry trade IIT, horizontal intra-industry trade HIIT, and vertical intra-industry trade VIIT during the period of 2000 to 2020, and also examine economic growth components impact on the IIT, HIIT, and VIIT. Estimated results examine that the bilateral trade relationship between China and (Afghanistan, Pakistan, and India) the strongly intra-trade and rapidly increased from 2000 to 2020, and intra-trade values between China and (Bangladesh, Maldives, and Sri Lanka) explain that the bilateral trade relationship between them decreased from 2000 to 2020, but Bhutan and Nepal intra-industry trade IIT increased but values are less than 15% which examine weak bilateral trade relationship between them, estimated results of VIIT explain that only Pakistan and India are two countries in South Asia region which have VIIT values positive which indicate that both countries have a two-way bilateral trade relationship with China, but the negative values of VIIT of other Six countries (Afghanistan, Bangladesh, Bhutan, Nepal, Maldives, and Sri Lanka) in South Asia region examine that they have one-way bilateral trade relationship with China.

The Hausman test results show that Fixed Effect Model is the best fit in the study when compared to Random Effect Model, but the Wald test reveals that Fixed Effect Model had Heteroskedasticity Problem, and the Wooldridge model results revealed that Fixed Effect Model did not have a Autocorrelation so as to be solved and that there is also no cross-section in the model, but to solve the Heteroskedasticity Problem we applied GLS model to test the Bilateral Economic Growth. The cross-section average GLS model estimates that Min. GDP, Avg. GDPPC and Language have a positive and significant impacts on intra-industry trade IIT but negative impacts on both HIIT and VIIT models, while LXMGDP has a negative impact on the intra-industry trade IIT but positive impacts on both HIIT and VIIT models. The GLS model was used by researchers to solve the heteroskedasticity problem in the models. According to the estimated results of the GLS cross-section difference method, the Min. GDP, Diff. GDPPC, Diff. FDI, Diff. GDP, Language, and L. Distance all have a positive and significant effects on IIT but have a negative effect on both HIIT and VIIT models, and LXMGDP has a negative effect on intra-industry trade but has a positive effect on both HIIT and VIIT models. The LXMGDP, Diff. FDI, and Max. GDP have negative effects on intra-industry trade IIT and negatively significant at 1% confidence interval on HIIT and VIIT, while the estimated results of Avg. FDI Diff. GDPPC, Min. GDP, and Language have positive and significant impacts on IIT but negative impacts on both HIIT and VIIT in the GLS model.

Policy Implications

Based on the findings of our analysis, we propose that China revise its export and import policies in order to reduce the volume of imports, as well as change its investment laws to make them more hospitable to investors in order to boost export productivity to South Asian countries.

Declarations

Note

*, ** and *** represent 1%, 5% and 10% significance level, respectively

Declarations

Note

*, ** and *** represent 1%, 5% and 10% significance level, respectively