Credit Development on China s Export

Transcription

1 Specialization or Diversification The Impact of Financial Credit Development on China s Export CHEN Xu QIU Bin (School of Economics and Management, Southeast University, Nanjing, , China) Abstract By using the data of Chinese manufacturing enterprises between 1999 and 2009, the specialization and diversification of China s exports are calculated. The relationship between financial development and the specialization of export and diversification of exports were studied by using dynamic panel data. It is found that financial credit development could significantly improve specialization of China s exports, while the diversification of exports is decreased by financial credit development. But there exsit significant differences among regions in China, Based on the empirical results, several policy suggestions are put forward to effectively improve the specialization and diversification of China s export of different regions. Keywords: financial credit, specialization of export, diversification of export, dynamic panel model JEL: F Introduction With the transformation of economic development model and increasingly open trade activities, China s export has showed a powerful growth in recent years regardless if volume or structure. In 1985, the proportion of manufactured products in China s total export was less than 50%. In 2012, its proportion increased dramatically to 95.1%. This indicates that China s export trade has successfully realized a transformation from primary products to manufactured products and reflected continuous improvement of China s export specialization level. Meanwhile, the diversification of China s export products is improving. Being calculated with UN-Comtrade by Qian Xuefeng and Xiong Ping (2010), about 1,009 types of commodities traded between China and 11 principal trade partners were increased during However, we have to admit that the trade development levels among Chinese regions 1

2 are significantly unbalanced. For example, it is found after handling the data in China Industry Business Performance Data that the export industries (4-digit) of industrial sectors in East China were 331 at average in By contrast, the export industries (4-digit) of industrial sectors in Central and West China were 165 and 92 respectively at average in The global economic recession caused by the subprime crisis in 2008 imposed a huge impact on foreign trade economy of various countries. This reminds us that we cannot neglect the important influences of financial credit environment on international trade activities. The positive roles of export specialization and diversification in regional economic growth and global value chain have been proved by some scholars (Liu Xiuyan and Wu Yan 2013: Yu Jinping and Deng Juan, 2014). Based on observation for the above facts, we cannot help thinking about the following: Will financial credit development be beneficial to China s export specialization or diversification as financial reform is increasingly fierce in China? From existing literatures, domestic and overseas scholars have conducted extensive and deep theoretical analysis and empirical studies on the relationship between export and financial credit development, but there are few investigations on the relationship between export structure and financial credit development based on export specialization and diversification. After reviewing lots of relevant literatures about financial credit development and export activities, we have reason to believe that there exists close relationship between financial credit and export specialization and diversification. The empirical findings for panel data of many countries applied by Beck (2003), Svaleryd and Vlachos (2005) indicated that if a country s financial market development could significantly promote the export expansion of manufactured products and particularly capital and technology-intensive products, the specialization or diversification level of manufactured products would be improved correspondingly. Then, Becker and Greenberg (2007) obtained the following similar conclusion based on an empirical study on the perspective of industrial heterogeneity in many countries: the progress of a country s financial credit system can effectively motivate the export behavior of the industries with high fixed export cost, and thus, will be conducive to the expansion of export sector types. Bilir and Davin (2014) also proved the restrictions imposed by strict financial credit system on a country s export. The analysis results for US monthly import data by them showed that the export trade volume of the countries with higher credit interest rates against USA declined significantly during global financial crisis. The radical decline of export volume means that its export specialization or diversification level is severely 2

3 weakened.the investment environment assessment results of the World Bank pointed out that China was one of the countries with the most severe financing restrictions. One of important causes for this situation is attributed to relatively backward credit environment and system. The empirical test for provincial panel data applied by Bao Qun and Zhang Yanan (2010) indicated that the development of Chinese regional finance could promote the formation of comparative advantages of hi-tech product export to some extent. It was found through theoretical model deduction and empirical test by Qi Junyan(2011) that the perfection of a country s financial system could promote the enhancement of export technical complexity by relieving adverse selection problems during production. Thus, the specialization of a country s export products could be improved. Similarly, the empirical results by Gu Guoda and Guo Aimei (2013) indicated that the financial development had significantly enhanced the complexity of China s export products. In addition to the investigation on the relationship between financial credit environment and export activities based on macro perspectives, the experimental evidences at micro level in recent years have also supported such a viewpoint that the improvement of financial credit system can effectively enhance enterprises export and influence specialization and diversification of regional export. The study results by Berthou (2010), Becker et al (2012) and Manova (2013) indicated that the perfection of financial credit system was conducive to lowering the cost threshold in export activities and realizing greater probabilities and volumes of export. Domestically, it has been an indisputable fact that Chinese private enterprises and SMEs are confronted with serious capital constraint during production and operation (Chen Lin et al. 2012; Han Jian and Chen Yan, 2014). Therefore, the overall improvement of financial credit environment can greatly promote the depth and breadth of Chinese enterprises export, which will be conducive to enhancing export specialization or diversification for those sectors restricted by finance. To sum up, it can be found that the influences of financial credit restriction at home and abroad on export are mainly based on export volume, export probability and technical content of export products. The direct relationship between export specialization and diversification and credit market is rarely mentioned. Under the real background that credit restriction can impose significant influences on China s export volume and structure, how the progress of financial credit market will influence export specialization and diversification? Under the real background that the levels of financial credit and trade development among Chinese regions are significant, what are the differences of the roles of credit 3

4 systems in various regions in export specialization and diversification? In view of this, dynamic panel models would be employed in this paper to investigate the problems above and propose corresponding policies and suggestions so as to enrich and deepen the study on the relationship between Chinese credit restriction and export trade. The roadmap of the paper is as follows. Section II discusses why financial credit development may play effects on the specialization and diversification of exports. Section III describes the data and empirical approach. Section IV presents the main results and robustness checks. Section V concludes. 2. Theoretical Mechanism Melitz (2003) is referenced in this paper to analyze the influences of external financing restriction on export involvement of production sector so as to accurately judge the relationship between the development of financial credit and export specialization and diversification. Taking Melitz (2003) as a reference, the sunk cost of production sector is f e, and its variable cost is q( ) /, in which represents productivity. Manova (2010) held that when the sector selected export, a certain iceberg cost ( 1) and prepaid fixed cost f will exist. Under financial restriction, suppose the sector has an x external financing proportion is e. That is, it finances ef from financial institutions like bank. To x obtain bank credit capital, the sector will mortgage or guarantee some tangible assets. Suppose the proportion for the sector to use the sunk cost is u(0 u 1) which is used as a guaranteeuf e. At present, the loan conditions provided by China s commercial bank to the enterprises request the enterprise to have sound financial information and other factors proving their loan repayment capacity, being represented by M. Therefore, e and u are the functions of M, and e ' ( M) 0, u ' ( M) 0. Under the above conditions, the influences of financial credit on export specialization and diversification levels can be analyzed in the following two situations. Situation 1: In the process of production, variable cost and fixed cost of export sector are all dependent on financial debit and credit. This situation is used to investigate the influences of financial credit on intensive margin of export sector. As the export sector gets involved in export trade further, the export specialization level of the 4

5 product or the industry will tend to be higher. At this moment, the optimal decision for export sector is: q( ) w max ( ) p( ) q( ) 1 e( M ) 1 e( M ) fx F( ) (1 ) u( M ) f (1) e p, q, F ( ) q( ) st.. A( ) p( ) q( ) 1 e( M ) 1 e( M ) fx F( ) q( ) B( ) e( M ) e( M ) fx F( ) (1 ) e( M ) fe 0 Wherein, A( ) and B( ) refer to export sector and restriction condition of financial institution. As a provider for external capital, the financial institutions like bank will obtain the revenue including interest from the export sector F( ). If the loan contract is executed with the probability ( 0 1), the export sector will pay financing cost F( ) for financing loan. Otherwise, the export sector will lose the guaranteed assetsuf. When the financing market is developed enough, the export sector will be e able to obtain financing loan easily. Suppose the external financing cost is 0, B( ) 0. At this moment, the export product price for export sector in case of maximum profit is p( ) / ( 1) (Melitz, 2003). refers to substitution elasticity among commodities. In case of financial credit restriction, if the productivity of export sector is less than the critical value of export and thus results in a failure to repay the financial credit capital, the export sector will need to increase export product price or decrease export scale so as to pay the cost and interest of credit capital. In the restriction conditions, A( ) F( ) : q( ) e( M ) 1 p( ) q( ) 1 e( M ) 1 e( M ) fx 1 e f x u( M ) fe, L is used to indicate the minimum productivity for export, and then export product price can be obtained: L em ( ) * p ( ) 1 e( M ) p ( ) 1 1 (2) It can be seen from Formula (2) that the export price of export sector under the restriction of 1 financial credit is higher. It can be judged from the export revenue r( ) R( ) 1P (Melitz, 2003) that the relationship between export volume and export price is a negative correlation. Thus, it can be judged that financial restriction has restricted the export volume of export sector and the export 5

6 specialization level. Situation 2: The sector s financial credit is used for fixed assets investment. This situation is used to investigate the influences of financial credit on export expansion margin of the sector. As the quantity and types of sectors involving in export trade increase, the regional export diversification level will be higher. At this moment, the condition for sector s profit maximization is shown in Formula (3): q( ) max ( ) p( ) q( ) 1 e( M ) fx F( ) (1 ) u( M ) f (3) e p, q, F ( ) : ' The critical productivity of export can be obtained by supposing ( ) 0 r ' ( ) 1 fx ( 1) e ( M ) fx u ( M ) fe (4) When the sector s productivity is ', export can be carried out. It can be seen from Formula (4) that compared with the situation without financial credit restriction (shown in Formula (2)), the critical productivity of export of the sector with financial credit restriction is improved. This indicates that financial restriction has restricted the possibility of the sector s export and the diversification of types of regional export sector. To sum up, the restriction of financial credit can not only reduce the sector s export probability but also impact some restrictions on the volume of export sector. Specialization and diversification of regional export are just caused by the types and export volume of export sector. Therefore, the influences of the development of financial credit on regional and national export specialization and diversification levels deserve further discussion. 3. Model Setting and Variable Description 3.1. Model Setting According to the action mechanism of financial credit for export behavior and by referring to the measurement model by Sun Shaoqin and Qiu Bin (2014), the empirical model in this paper is set as below: extrade fasupport termstructure finsize open i, t 0 1 i, t 2 i, t 3 i, t 4 i, t 6

7 5 invest i, t 6 tech i, t 7 road i, t 8 gov i, t i, t i, t Wherein, the structure extrade of export trade and explained variable includes h, hdiv and vdiv, representing province i s export specialization, export horizontal diversification and export vertical diversification in t year. Explaining variables fasupport, termstructure and finsize, represent a financial institution s credit support in province i for fixed assets investment, credit term structure and credit amount against the volume of gross regional production in t year. Meanwhile, several control variables are also set in this paper, whereinvest is about the fixed investment in the region so as to control the influences from capital accumulation. tech reflects R&D innovation capacity of each province. In addition, some common variables reflecting policies, geography and system, including open, road and gov. open represents economic openness in the region, it road reflects transportation in the region, and gov reflects the economic intervention by government Variable Measurement Explained Variables Export specialization index ( h ): At present, HHI is used to measure export specialization in most of the literatures (Liu Xiuyan, 2013; Shao Jun, 2011). It is a comprehensive index which is used to measure industrial concentration. After that, it is gradually used to measure distribution characteristics of a certain variable, and its value goes between 0~1. Taking the practice by Liu Xiuyan and Wu Yan (2013), the following formula is applied to calculate HHI. h n x 2 ijt j X it 1 1 N it 1 Nit In this formula, region i in t year, x ijt refers to export value of industry j of X it refers to gross export of industry of region i in t year, N refers to quantity of export industries (4-digit) of region i in t year. If h is closer to 0, it it indicates that the degree of export specialization will be lower. If h is closer to 1, it indicates that the degree of export specialization will be higher. In particular, industry export amount and regional export 7

8 amount are obtained by totaling and systemizing China Industry Business Performance Data during Export diversification index: Taking Herzer and Nowak-Lehnmann (2006) as a reference in this paper, export horizontal diversification ( hdiv ) and export vertical diversification vdivit, are measured by natural logarithm of 4-digit industry quantity of regional export and the proportion of manufacturing export in gross export in every province Explaining Variables Development index of financial credit: Under the real background that Chinese credit system is based on bank and considering the availability of regional credit data, the development speed of regional financial credit market is measured by a financial institution s support of fixed assets investment, credit term structure and credit amount against the volume of gross regional production so as to fully observe its influences on export specialization and diversification. Credit support of fixed assets investment ( fasupport ): Compared with non-export enterprises, export enterprises are also confronted with high export fixed cost. The difficulty of credit of fixed assets investment capital might influence export decision, which will change the export structure in the region. Therefore, this index is measured by domestic loan proportion of fixed assets investment in every province. As the proportion is higher, it indicates that the credit development of the region is higher. Term structure of ( termstructure ) of financial credit: Compared with short-term loan, medium and long term loan can help strengthen enterprises risk tolerance at a long production and operation period. Thus, this might influence export specialization and diversification levels in Chinese provinces. Therefore, the proportion of medium and long term loan in total loans of financial institutions in every province will serve as an index to reflect the term structure of financial credit. Financial credit scale ( finsize ): The financial credit scale of every province is measured by the proportion of total loans from regional financial institutions in GDP. If the proportion is higher, it indicates that the overall scale of regional financial credit will be greater. Economic opening level ( open ): Regional economic opening level is one of the important factors 8

9 to influence export structure. Regional economic opening level is measured by the proportion of actually utilized amount of foreign direct investment in GDP. Fixed assets investment ( invest ): With further development of Chinese industrialization and opening to the outside world, fixed assets investment makes increasing contribution to economic growth and export trade. Regional fixed assets investment is measured by the proportion of fixed assets investment of every province in GDP. R&D innovation level ( tech ): Innovation capability is one of important sources to enhance the industrial added value of export products. With improved R&D capacity, export structure (including export specialization and export diversification) will also have certain changes. R&D innovation capacity is measured by the quantity of domestically applied and authorized patents in every province. Its natural logarithm is adopted to eliminate the influence of heteroskedasticity. Traffic facilities ( road ): Complete basic traffic facilities will facilitate the flow of elements and HR, which changes the element structure in the region to some extent and influences the export s comparative advantages. The mileage of roads and railways per capita in every province is used as a proxy variable for basic traffic facilities. Economic activity intervention by government ( gov ): Compared with those countries and regions with high marketization degree, one of important characteristics for Chinese socialist market economy is that it has high government invention. To investigate the influence of every government s economic intervention on export specialization and export diversification, the proportion of general fiscal expenditure in GDP is introduced as an index. The above explaining variables are sourced from annual China Statistical Yearbook and Chinese Finance Statistical Yearbook. Export specialization of explained variables is calculated through China Industry Business Performance Data during Taking the differences of statistical caliber and data loss into consideration, the data relating Hong Kong, Macau and Tibet etc is included as a study sample. Finally, the study samples include the panel data of 30 Chinese provinces during Table 1 provides a simple statistical description for the variables in national samples. It can be seen that the export specialization level in the west of China is significantly higher than the east and central parts. The possible reason is that natural resources and a majority of product types like agricultural 9

10 products in the west of China dominate export trade, so HHI in the west is relatively high. However, the high-level export specialization implies the west of China is excessively dependent on primary exporting to obtain short-term economic benefits, suffering a risk of resource curse. In export horizontal diversification and vertical diversification, export product types and structure in the east of China take a leading position in China, being followed by the central part and the west of China. Besides, credit support of fixed assets investment by financial institutions in the west of China is equivalent to the east of China but it is significantly higher that the central part of China. This might benefit from the capital support for the west region in the west development strategy, which contributes a high proportion of bank loan in fixed assets investment in the west of China. In financial credit structure and credit scale, the west region keeps ahead of the central region, flowing after the east of China. Generally, the financial credit development in the east of China takes the leading position in China, being followed by the west region. The financial credit level in the central region lags behind relatively. 4. Empirical Result Analysis To overcome endogeneity problems among variables, system GMM estimation put forwarded by Arellano and Bover (1995) is adopted in this paper. Specifically, when using the system GMM estimates the variables, the instrumental variables are obtained after first difference of variables, and the lagging items of first difference of variables are treated as instrumental variables according to Arellano and Bond (1991). Thus, the endogeneity among variables can maximally overcome. To avoid colinearity influences among different financial credit indexes, variables include of fixed assets investment credit support, term structure and credit volume are regressed with the explaining variable respectively. Moreover, the influence differences of credit development on export specialization and diversification at different perspectives can be compared Nationwide Regression Analysis Table 2 exhibits the regression results based on nationwide from system GMM estimation. In particular, the explained variable in (1)-(3) is export specialization, and the explained variables in (4)-(6) and (7)-(9) are horizontal diversification and vertical diversification respectively. The regression coefficient of credit support of fixed assets investment is and has passed significant testing for 1%. 10

11 This indicates that the increase of domestic loan proportion in fixed assets investment will be helpful to improve China s export specialization. Similarly, the increase of medium and long term loan displayed by the regression coefficient of term structure of financial credit and financial credit volume index as well as total loan increase can significantly enhance China s export specialization to strengthen China s export trade competiveness. Compared with financial credit volume index, the estimated coefficient of the credit support of fixed assets investment and credit term structure have higher values. Therefore, export activity can be effectively promoted by supplying the capital of fixed assets investment and increasing medium and long term loan. Especially for manufacturing with high-tech features, export specialization level can be enhanced by learning by doing and technology spillover effect. This also indicates that attention should be paid to optimization of credit efficiency and structure while the financial credit volume is expanded. Compared with export specialization, China s export diversification level shows a certain decline as financial credit develops. In export horizontal diversification, the term structure of financial credit and the estimated coefficients are negative and significant at 1%. The index regression coefficient of the support of financial credit for fixed assets investment is also negative, but they fail to pass significance testing. Such results might be attributed to China s credit capital used in specialized production. To avoid the risks arising from overseas market during new product development, export sectors will usually invest the obtained capital on original export products. This will make the types of China s export sectors tend to be concentrated. In estimated results for export vertical diversification, the estimated coefficients of credit support of fixed assets investment and credit structure index are positive, but it fails to pass significance testing. This indicates that the utilization of fixed assets investment capital and medium and long term loan has not fully performed their roles in improving the proportion of manufacturing in export. It is worth noting that the increase of China s financial credit volume has significantly restricted export vertical diversification. This result might be attributed to the following: the allocation efficiency of loan capital is low or the loan enterprises invest more capital on the domestic market. This indicates that to simply expand loan scale cannot effective improve China s export vertical diversification. Attention should be paid to reasonable allocation for credit capital and optimization of credit term structure. Meanwhile, the regression result of control variable also provides other important information. 11

12 Firstly, the increase of fixed assets investment scale can lower export specialization level to some extent and improve export diversification level. This result might be attributed to the following: the increase of fixed assets investment is more utilized to increase the types of export products and optimize export product structure. This has increased the expansion margin of China s export, so it plays a restricting role in specialization. Secondly, the increase of economic openness degree produces completely opposite influences on export specialization and export horizontal diversification: the expansion of foreign capital utilization scale can significantly increase the types of export sectors but significantly lower export specialization level at the same time. Besides, R&D innovation capacity can significantly promote export horizontal diversification, which indicates that technical innovation is a key factor in increasing China s export expansion margin. Finally, the different influences of government expenditure on export specialization and diversification might be attributed to the differences of the targets in financial support Regression Analysis of Region Division Financial Credit Development and Export Specialization To further research the influences and differences of financial credit development in different Chinese regions on export specialization level. In this paper, the samples are divided into east region, central region and west region. The regression results are shown in Table 3. In the east and center of China, the estimated coefficients of credit support of fixed assets investment, credit term structure and credit volume are significantly positive, which indicates that the improvement of overall development level of financial credit in the east and center of China can significantly improve export specialization level. At present, the upgrading of product structure from primary products to manufactured products has been basically realized in the east of China, so the progress of financial credit market has strengthened the export competitiveness in the east region. The export trade in the center of China is possibly under an upgrading stage from primary products and low value-added products to manufactured products with high value content, so the proper inclination of financial resources to high and new technology industries might accelerate the optimization and upgrading of export structure in the central region. The research results by Liu Xiuyan and Wu Yan (2013) indicate that the improvement of export specialization can promote the economic growth in the east and center of China to some extent, so the promoting role of financial credit development in export specialization should be played continually in the east and center 12

13 of China. By contract, the influences of three indexes to measure financial credit development on export specialization in the west of China fails to pass significant testing. The influences of financial credit development in the east and central of China are higher than the west region. It is held in this paper that this result is attributed to the following: compared with the west region, the capital marketization degree in the east and center of China is relatively higher, the credit system is relatively perfect, and export enterprises can obtain the capital for production and operation more efficiently. The improvement of market system in the west region is far beyond the east and west regions. The shortage of efficient financing channels makes the limited capital fail to be fully utilized. In fact, the export specialization in the west of China relies more on local resource endowment to establish comparative advantages. While improving allocation efficiency of financial resource, attention is also paid to transfer and development of export trade from primary products to manufactured products so as to prevent resource curse. Compared with the east and center of China, the financial credit service and credit system environment in the west of China lag behind significantly, which can hardly provide a powerful capital support for export trade and particularly expert specialization production with a certain technical content. Meanwhile, the influences of control variable on export specialization have some differences among different regions. For example, the degree of economic openness can significantly restrict export specialization in the east and center regions. The estimated result in the west region is positive, but it fails to pass significant test. It might be attributed to the following: the further deepening of the degree of economic openness in the east and center of China and the effect of technology spillover of multinational companies have stimulated development and production of new products. In the west of China with the lowest economic openness level, the entering of foreign capital does not produce a significant influence on export specialization. With the increase of the proportion of investment in fixed assets, the export specialization in the east region shows a significant decline trend. The reason for this result may be that in the east region which brought together a large number of export enterprises, investment in fixed assets are more used to overcome the fixed costs of export by new export sectors, and thus, export specialization has been inhibited by fixed asset investment. Compared with the center and west region, the improvement of traffic facilities in the east region does not produce significant influences on export specialization. It is worth noting that the influences of government s economic intervention on export specialization play a completely opposite role in the east and west of China. This might be attributed to 13

14 the differences of economic behaviors of local governments Financial Credit Development and Horizontal Diversification Export diversification can be divided into export horizontal diversification and export vertical diversification. In particular, the improvement of export horizontal diversification will be helpful to resist the influences of fluctuations of international economic environment on export trade. The depending of export vertical diversification is, to some extent, a reflection of production efficiency of export products and structure of export products in manufacturing, which will be helpful to enhance the trade competiveness. Refer to Table 4 for the influences of Chinese regional financial credit level on the types of export sectors. It can be seen that in terms of the impact imposed by financial credit development on export horizontal diversification, there exists significant differences among regions in China. Export horizontal diversification shows a significant decline tendency with the increase of financial credit in central region. In addition to the coefficient credit support for investment in fixed asset fails to pass the significance test, there are significant negative correlation between financial credit development and export horizontal diversification on the whole. By combining Table 4, this result can be interpreted as more credit capital being used in export specialization production, which reduces enterprises capital investment on new product development. But in contrast, the estimated coefficients of export horizontal diversification in the west region against different indexes to measure the development of financial credit fail to pass significance test. It might be attributed to the following: on the one hand, the credit system and environment construction in the west of China are relatively backward. The low efficiency of financial resource allocation results in the limited functions of credit market for regional export diversification. On the other hand, the types of export industries are relatively simplex, and export trade is based on primary products with a shortage of technical content. The credit resources obtained by the enterprise are used in specialized production. The estimated coefficient of west region in Table 3 can reflect this fact. The regression results of control variables in Table 4 provide some additional valuable information. In degree of economic openness, the export horizontal diversification in the east region can produce a significant promoting role, but this influence is insignificant in the center and west regions. This might be caused by the differences of foreign capital utilization focus and efficiency. The increase of fixed 14

15 assets investment in the east region can significantly enrich the types of export products and optimize export structure. The west region goes the opposite way. This might be attributed to the following: the export trade types in Chinese west region are focused on simplex primary products, and the fixed assets investment for export trade is more used to strengthen the original primary product production and export, which has restricted the expansion of export product types. The positive role of R&D innovation capacity in export horizontal diversification is reflected in the east, center and west of China. Opposite to the results in Table 4, the improvement of traffic facilities has significantly restricted export horizontal diversification in the center and west of China. This might be attributed to the differences of the flow of production elements and the changes of element structure proportion in production influences after the improvement of traffic convenience. Compared with export specialization, the export horizontal diversification in the east and west of China is significantly and negatively influenced by the financial expenditure of local governments Financial Credit Development and Vertical Diversification In addition,this paper further investigate the regional differences for the influences of financial credit development on export diversification, The results are shown in Table 5. The financial credit development level in the east region has restricted export vertical diversification to different extents. The estimated coefficient of financial credit volume is and significant at 5%. This indicates that the expansion of financial credit scale in the east of China has restricted the increase of the proportion of export trade in manufacturing. The credit support of fixed assets investment and the term structure of financial credit have an insignificantly negative correlation with export vertical diversification. This shows that the development of financial credit does not promote the manufacturing proportion in export trade in the east region. This result might be attributed to the following: on the one hand, the east of China is always a region with the highest manufacturing density and export specialization level in China, and the credit capital might be more used in production of export specialization. On the other hand, the east of China has more financial networks and mature credit service system, and the manufacturing has gone through a relatively long expansion stage. With the rapid growth of service trade and other trades, financial credit capital is more distributed in emerging industries, which has restricted the enhancement of traditional manufacturing export. Compared with export horizontal diversification, the regression 15

16 coefficient of financial credit in the center region has a significant change. The proportion of fixed assets loan and ratio of medium and long term loan can produce an effective promoting role in export vertical diversification in the center region. The estimated coefficient of financial credit volume fails to pass significance test. This indicates that the support of financial institutions in the center of China for medium and long term investment will help enhance the export vertical diversification level. Similar to the results in Table 3 and Table 4, the financial credit development in the west of China does not play a significant role in improving export vertical diversification. This might be attributed to the relatively backward financial service system and small manufacturing export volume in the west region Robustness Test To further test the robustness of above empirical results, FGLS approach is applied to estimate the samples. The results are shown in Table 6. To save the space, the table only exhibits the estimated results of core explaining variables and places the estimated results of financial credit indexes at different levels in the same column. It can be seen from the aforesaid empirical results that the positive and negative directions and significance of estimated coefficients in the samples of Table 6 are basically consistent with the estimated results of the system GMM. For example, it can be found from a nationwide perspective that the capital support of Chinese financial system for fixed assets investment, the increase of the proportion of medium and long term loan and the improvement of financial loan volume can significantly strengthen export specialization level. By a sharp contrast, the types of export sectors show a shrinking trend as the financial credit volume and the proportion of medium and long term loan increases. Meanwhile, the expansion of overall volume of financial credit has significantly restricted the proportion of manufacturing export in export trade. The estimated results by division are not stated here. Therefore, the empirical results in this paper are of robustness. 5. Conclusions and Policy Suggestions Under the trend and the real background that the relationship between financial and international trade activities are increasingly close, the influences of financial credit development on export trade are reflected in export volume and export product structure. With the important influences of export specialization and diversification on export structure and economic growth, the influences of financial 16

17 credit development on regional export specialization and diversification have been a research focus in this paper. The empirical results show that there are differences for the influences of different indexes to measure financial credit development on export specialization and diversification. Firstly, from nationwide level, the development of financial credit at different levels can significantly strengthen China s export specialization level. By a sharp contrast, export horizontal diversification shows a decline as the financial credit develops. Meanwhile, the expansion of financial credit volume has significantly reduced China s export vertical diversification level. Further regional research results indicate that: (1) the progress of financial credit system can significantly improve the export specialization level in the east and center of China. This function is not significant enough in the west of China; (2) the export specialization and financial credit development show a significant positive correlation in the east and center of China, while the correlation is negative in terms of export horizontal diversification. (3) financial credit development significantly promotes the vertical diversification in the central region of China, while the vertical diversification has been inhibited by the progress of financial credit. (4) the relationship between financial credit development and export specialization and diversification fail to pass significance test in the west of China. To sum up, the following policy suggestions are proposed in this paper: Firstly, the east of China is usually a dense region for manufacturing product export with high technical content. To fully play the promoting role of the progress of financial credit system in specialized production is an important way to enhance the status of global value chain of China s export products. While the financial credit fully plays its role in export specialization level in the west of China. Attention should be paid to the transformation and upgrading of export products from primary products to manufactured products so as to avoid the comparative advantage trap. Secondly, it is pointed out by Liu Xiuyan and Wu Yan (2013) that export diversification plays a positive promoting role in stability of China s export trade and regional economic growth. In every region and particularly the east region, attention should be paid to the capital support for fixed assets investment in export production, and the financial resources are properly utilized in emerging manufacturing export operation and manufacturing export expansion margin. The export diversification can be enhanced by allocating financial resources more effectively so that good demonstration effect can be formed in the center and west of China. Besides, some financial resources should be properly inclined to emerging export industries and industrial manufacturing in every region to 17

18 strengthen export specialization and keep improving export diversification level. Finally, proper financial reform and perfection should be made according to the relationships among export specialization, diversification and financial credit development in every region. Proper adjustment should be made for multiple economic behaviors and systems which can influence regional export specialization and diversification according to specific export trade development plan, degree of economic openness, fixed assets investment and government s economic intervention etc. 18

19 References Arellano M. and Bover O.. Another Look at the Instrumental Variable Estimation of Error-components Models. Journal of Econometrics, 1995, 68:29~51. Arellano M. and Bond S.. Some Tests of Specification for Panel Data: Monte-Carlo Evidence and an Application to Employment Equations. Review of Economic Studies, 1991, 58: 277~297. Bao Qun and Zhang Yanan., Financial Development Comparative Advantage and High-tech Export of China, Studies of International Finance No.11, 2010, pp Becker B., Jinzhu Chen and David Greenbreg., Financial Development, Fixed Cost, and International Trade. The Society for Financial Studies. Oxford University Press, Beck, T.,Financial Development and International Trade: Is There a Link. Journal of International Economics, No. 5, 2002, pp Berthou, A., The Distorted Effect of Financial Development on International Trade Flows.CEPII Working Paper, 2010, No. 9. Bilir L. K., Chor D. and Manova K., Host-Country Financial Development and Multinational Activity. National Bureau of Economic Research Working Paper, Chen Lin. He Huanlang, Luo Changyuan, Financing Constrains and Export Behavior of SMEs: Extensive and Intensive Margins,Journal of Finance and Economics, No. 10, 2012, pp Gu Guoda, and Guo Aimei., Financial Development and Export Complexity Increase:An Empirical Test Based on Functioning paths, International Economics and Trade Research,No.11, 2013, pp Han Jian and Chen Yan, Financial Development and the binary marginal of export enterprises, Forum of World Economics & Politics, No.1, 2014, pp Herzer D., Nowak- Lehnmann F.. What Does Export Diversification do for Growth? An Econometric Analysis. Applied Economics, 2006, 38(15):1825~1838. Kalina Manova., Credit Constraints, Heterogeneous Firms, and International Trade. Review of Economic Studies,,No. 80, 2013, pp Keller W., International Technology Diffusion. Journal of Economic Literature, Vol. 42, No. 3, 2004, pp Kletzer K. and P.Bandhan., Credit markets and Patterns of International trade. Journal of Development Economics, 1987,pp Javorcik B. S. and Spatareanu M., Does it matter where you come from? Vertical spillovers from foreign direct investment and the origin of investors. Journal of Development Economics, Vol. 96, No. 1, 2011, pp Liu Xiuyan. and Wu Yan, The Specialization of Export, the Diversification of Export, and the Growth of the Regional Economy: A Case Study Based on the Panel Data of China s Provinces and Cities of the Provincial levels,management World, No.8,2013, pp Qi Junyan, Financial Development and Competitiveness of Trade: A Panel Data Analysis based on China s Industries, World Economy Study, No.5, 2010, pp Qi Junyan, Wang Yongjin,Shi Bingzhan, Sheng Dan, Financial Development and Export Technical 19

20 Complexity, The Journal of World Economy, No.7, 2011, pp Qian Xuefeng and Xiongpin, The Dual Margin of China Export Growth and Its Determinants, Economic Research Journal, No.1, 2010, pp Sun Shaoqin and Qiu Bin, Research on the Relationship between Financial Development and China s Optimizing Export Structure Based on the Perspective of Regional Differences, NanKai Economic Studies, No.4, 2014, pp Svaleryd H. and Vlachos J., Financial Markets, the Pattern of Specialization and Comparative Advantage: Evidence from OECD Countries. European Economic Review, No. 49, 2005, pp Wynne, J., Wealth as a Determinant of Comparative Advantage. The American Economic Review, Vol. 95, No. 1, 2005, pp Xu Jianjun and Wang Haohan, The Effects of Financial Development on Foreign Trade and Their Regional Difference in China Cointegration Analysis and GMM Based on Crossed Provinces Panel Data, Journal of International Trade, No.4, 2008, pp Yu Jingping and Deng Juan, Vertical Specialization, Technology of export and Global Value Chain Division. Forum of World Economics & Politics, No.2, 2014, pp

21 Appendix Table 1: Variable descriptive statistics Core variable Nationwide East Part Central Part West Part Export specialization Export horizontal diversification Export vertical diversification Credit support of fixed assets investment Term structure of credit Credit scale Data source: The data is collected and calculated in this study. The values in the table are mean values. 21

22 Table 2: Development of financial credit and export specialization and diversification: System GMM estimation nationwide Export specialization horizontal diversification Vertical diversification (1) (2) (3) (4) (5) (6) (7) (8) (9) fasupport *** (0.214) (0.525) (0.497) termstructure *** *** (0.151) (0.261) (0.333) finsize *** *** *** (0.058) (0.117) (0.073) open * * ** *** ** (0.357) (0.449) (0.393) (1.895) (1.889) (1.997) (0.914) (0.831) (0.702) invest * ** ** * * *** *** *** (0.055) (0.080) (0.094) (0.418) (0.363) (0.317) (0.246) (0.232) (0.180) tech *** *** *** *** *** (0.014) (0.017) (0.011) (0.064) (0.064) (0.070) (0.022) (0.035) (0.023) road * *** *** *** ** (0.001) (0.001) (0.001) (0.003) (0.003) (0.003) (0.003) (0.002) (0.003) gov * *** *** * *** *** ** (0.220) (0.303) (0.273) (0.758) (0.657) (0.770) (0.679) (0.673) (0.549) AR(1) AR(2) hansen test Year Control Control Control Control Control Control Control Control Control Obs Note: ***, ** and * represent 1%, 5% and 10% significance levels. The values in bracket are stable standard deviations. 22