Downloaded from: Publisher: Business Perspectives. Please cite the published version.

Transcription

1 Abdou, HAH and Wang, Y and Li, X and Nim, C (205)Financial developmen and economic growh in China. Invesmen Managemen and Financial Innovaions, 2 (3). pp ISSN Downloaded from: hp://e-space.mmu.ac.uk/68244/ Publisher: Business Perspecives Please cie he published version hp://e-space.mmu.ac.uk

2 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 Yan Wang (UK), Xiaoyu Li (UK), Hussein A. Abdou (UK), Collins G. Nim (UK) Financial developmen and economic growh in China Absrac The purpose of his paper is o examine he relaionship beween financial developmen and economic growh. In paricular, he auhors examine he impac of financial developmen on he growh of primary, secondary, and eriary indusries in China. Ordinary Leas Square (OLS) muliple regressions are applied on a se of daa from China for he period 978 o 203 o deermine he effecs of financial developmen on economic growh, while conrolling for oher macroeconomic variables, namely labor force, capial growh, inflaion rae and expor growh. The empirical resuls show ha financial developmen has a negaive effec on economic growh in general, bu on he growh of he eriary indusry in paricular. By conras, i is found ha financial developmen has no significan effec on he primary and secondary indusries. The findings offer policymakers some useful insighs ha more aenion may need o be paid on developing capial markes and providing more invesmen choices/opporuniies for Chinese households. This paper is differen from mos of he previous sudies as i uses up-o-dae daa ( ) from China capuring he effecs of financial developmen on economic growh in addiion o oher macroeconomic facors. Keywords: economic growh, financial developmen, China, hree major indusries (primary, secondary and eriary indusry). JEL Classificaion: N, O, O43. Inroducion The cenral objecive of his paper is o invesigae he impac of financial developmen on economic growh in China. Specifically, i seeks o measure he effec ha financial developmen has on he growh of he primary, secondary and eriary indusries. Since 973, he heory of financial liberalizaion has been carried ou by Mckinnon (973) and Shaw (973). They argue ha he liberalized financial secor or developed financial secor can simulae economic growh by efficienly allocaing resources. Since he idea of financial liberalizaion has been recognized by more and more policymakers, many counries have begun o reform heir economies in order o develop he financial secor. In 978, Chinese governmen launched an economic reform in order o ransform he planned economy (pre-978) o marke economy (pos-978). Since hen, he economic developmen in China has experienced remendous changes, many enerprises have been esablished during ha period, and he local governmens have gained more powers and righs o allocae resources (Naughon, 2007). Yueh (203) suggess ha he pos-978 economic reform has led o high economic growh in China. China has begun o experience high GDP growh rae, and he average annual growh rae of GDP is around 0% even in he period of he 2007/2008 financial crisis. Yan Wang, Xiaoyu Li, Hussein A. Abdou, Collins G. Nim, 205. Yan Wang, Deparmen of Accounancy and Finance, Universiy of Huddersfield Business School, Universiy of Huddersfield, UK. Xiaoyu Li, Deparmen of Accounancy and Finance, Universiy of Huddersfield Business School, Universiy of Huddersfield, UK. Hussein A. Abdou, Huddersfield Business School, Universiy of Huddersfield, Huddersfield, Wes Yorkshire, UK, HD 3DH; Managemen Deparmen, Faculy of Commerce, Universiy of Mansoura, Mansoura, Dakahlia, Egyp. Collins G. Nim, Deparmen of Accounancy and Finance, Universiy of Huddersfield Business School, Universiy of Huddersfield, UK. 8 In 20, he oal GDP in China reached 5.93 rillion US dollars, wih China overaking Japan and becoming he second larges economy in he world (Naional Bureau of Saisics of China, 203). Some researchers have invesigaed he relaionship beween financial developmen and economic growh in China. Zheng and Yu (2009) find ha here is a posiive relaionship beween financial developmen and economic growh by using he annual daa of 29 provinces over he period 994 and This resul is suppored by Wen (2009) who uses daa from he cenral region (cover six provinces) of China over he period 978 o By using he annual daa from 286 ciies over he period , Zhang e al. (202) also find ou ha here is a posiive relaionship beween financial developmen and economic growh. In conras, Hasan e al. (2009) find ha here is a negaive relaionship beween financial developmen and economic growh by using annual daa of 3 provinces over he period More recenly, by using he annual daa of 34 provinces for China over , Fang and Jiang (204) examine he effec of financial developmen on primary, secondary, and eriary indusries in China. They find ha financial developmen has no effec on he growh of primary indusry, bu has a posiive effec on he growh of secondary and eriary indusries. Differen from mos previous sudies, which used provincial daa or ciy-level daa o examine he relaionship beween financial developmen and economic growh in China, his paper firs examines he relaionship beween financial developmen and economic growh in general, and hen, respecively, examines he relaionship beween financial developmen and he growh of he primary, secondary, and eriary indusries using counry level daa, which covers he period beween 979 and 203.

3 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 The aim of his paper, herefore, is o examine he relaionship beween financial developmen and economic growh in China by using counry level daa over he period More specifically, we aemp o answer he following quesions. Firsly, wha kind of effecs (i.e., posiive/ negaive/no) does financial developmen has on economic growh in China? Secondly, among he hree major indusries (primary, secondary, and eriary indusry), which indusry has mosly been influenced by financial developmen in China? By using mulivariae regression models, we find ha financial developmen has a negaive effec on economic growh and he growh of he eriary indusry. In conras our findings sugges ha financial developmen has no significan effec on he growh of he primary and secondary indusries over he period The res of his paper is organized as follows. Secion one reviews he relevan lieraure. Secion wo describes he sample and research design. Secion hree presens he empirical resuls. The final secion concludes he sudy and suggess areas of fuure research.. Lieraure review A number of empirical sudies has been underaken o examine he relaionship beween finance and growh, as discussed below. Mos of hem are based on endogenous growh heory, which demonsraes ha economic growh can coninuously increase because of endogenous forces, such as echnological progress, human capial accumulaion, research and developmen (i.e., R&D). In his paper, reviewing he relevan lieraure based on cross-counry sudies and counry-specific sudies is underaken. Cross-counry sudies: hese sudies mainly use panel daase for many counries o examine he finance-growh nexus. The findings of one of he early cross-counry sudies underaken by King and Levine (993) suppored he views of Schumpeer (9), who believed ha a developed financial sysem can effecively allocae saving o invesmen, and hus simulae economic growh. King and Levine (993) analyze panel daa for 80 counries over he period , and find ha financial developmen is posiive and srongly relaed o economic growh. They sae ha a developed financial sysem can improve he efficiency of invesmen and increase capial accumulaion, and hus simulae economic growh. Gregorio and Guidoi (995) find similar resuls o hose by King and Levine (993). They use panel daa for 98 counries during , and conclude ha financial developmen is posiively correlaed o economic growh, as financial sysem can make a more efficien invesmen. Moreover, hey creaed anoher panel daase for 2 counries of Lain America over he period , and find ha here is no relaionship beween financial developmen and economic growh. Gregorio and Guidoi (995) argue ha hose Lain American counries pursued financial liberalizaion reforms wihin relaively poor regulaory environmen, and hus i seems harder for financial developmen o play significan posiive role in economic growh. Since hen, many scholars realized ha he financegrowh relaionship may vary depending on he specific characerisic of an economy, such as he level of economic developmen and income. Thus, a considerable number of sudies have been done whose sample selecion has been influenced by he characerisic of he counry or counries under sudy. By analyzing daa for 30 developing counries during he period , Al-Yousif (2002) saes ha here is a bi-direcional causal relaionship beween financial developmen and economic growh. Apergis e al. (2007) analyze daa for 5 OECD counries and 50 Non-OECD counries over he period They argue ha he policies for simulaing financial developmen also have a posiive effec on economic growh and vice versa. Moreover, using a panel daa for 27 ransiion economies, which have ransformed from planned-economy o markeeconomy, over he period , Akinov e al. (2009) found ha here is a significan posiive correlaion beween financial developmen and economic growh. They argue ha hose counries have already experienced long-erm ransiion of heir economies from planned-economies o a markeeconomies, and hus heir financial sysem can be more efficien in allocaing resources based on he supply and demand of goods (Akinov e al., 2009). Handa and Khan (2008) choose 3 counries o evaluae wheher he causaliy relaionship relaes o he differen sages of economic developmen using daa from 960 o These 3 counries include four low-income counries, namely Bangladesh, India, Pakisan, and Sri Lanka; five middle-income counries, namely Argenina, Brazil, Malaysia, Thailand and Turkey; and four high-income counries, namely Germany, Japan, UK and USA. They find ha India and four high-income counries have a bi-direcional causal relaionship beween financial developmen and economic growh, and oher low-income and middle-income counries have unidirecional causaliy running from economic growh o financial developmen (i.e., OECD refers o Organizaion for Economic Co-operaion and Developmen, which is composed of 34 counries wih a marke economy. 9

4 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 demand-following hypohesis). Handa and Khan (2008) argue ha because he financial sysems in low-and middle-income counries end no o be, as developed as hose of high-income counries, and herefore he posiive role of he financial sysem on economic growh is no easily observed. Moreover, and alhough India belongs o low-income counries; is financial secor is relaively developed, wih he counry boosing of a variey of financial insiuions and insrumens. Thus, India s financial sysem can help in improving he efficiency of invesmen, and simulae economic growh beer han he oher low-income counries. Mos of he previous sudies show ha financial developmen could eiher have posiive influence or no influence on economic growh. However, he 2007/2008 financial crisis demonsraed ha sophisicaed financial sysems can also someimes have negaive effecs on economic growh. Law and Singh (204) explain ha he failure of he financial sysem can lead o a wase of resources, a decrease in savings and an increase in speculaion, which can cause problems of misallocaion of resources and underinvesmen. As a resul, he economy can hus conribue owards increasing unemploymen and povery raes. Therefore, some scholars argue ha here should be a hreshold relaionship beween financial developmen and economic growh, which means ha financial developmen can have posiive impac on economic growh a cerain hresholds, bu no or even negaive effecs a oher hresholds. Arcand e al. (202) analyze panel daa for 00 counries during They find ha when he raio of privae secor credi o GDP is below 00%, financial developmen ends o simulae economic growh; conversely, when ha raio exceeds 00%, furher developmen of he financial secor ends o have negaive effec on economic growh. Law and Singh (204) also provide evidence on he hreshold relaionship beween finance and economic growh using panel daa for 87 counries over he period 980 o 200. They show ha he hreshold value is 88% when he proxy of financial developmen is he raio of privae secor credi o GDP, and he hreshold value is 9% when he proxy of financial developmen is he raio of illiquidiy liabiliy o GDP. This means ha finance can have a negaive effec on economic growh when he raio of credi o he privae secor exceeds 88%, or when he raio of illiquidiy liabiliy of GDP exceeds 9% (Law and Singh, 204). Counry-specific sudies: alhough he Crosscounry sudies give an overall view of he financegrowh relaionship, i ignores he specific characerisic of each counry. Therefore, some scholars who are ineresed in he finance-growh relaionship of specific counry would like o analyze ime-series daa for one counry. Ghali (999) invesigaes he finance-growh relaionship in Tunisia using annual daa over he period He finds ha financial developmen can lead o economic growh in Tunisia. Moreover, he financial secor is sill underdeveloped in Tunisia, and hus furher sudies should be conduced by he governmen of Tunisia o furher develop he financial secor, such as liberalizing ineres rae and increasing he availabiliy of differen ypes of financial insrumens (Ghali, 999). Differen from Tunisia, finance developmen does no have much influence on economic growh in Tanzania. By analyzing annual daa for Tanzania during , Akinboade (2000) find ha financial developmen has no influence on economic growh during , and hus has lile posiive influence on economic growh during Tha is because he financial sysem does no operae efficienly over he period , as i is mainly conrolled by he governmen; and from 980s, he financial sysem became more liberalized because of he pursuance of financial reform. Akinboade (2000) argues ha financial developmen has a posiive influence on economic growh since he financial reform in Tanzania has aken roo. Lee and Wong (2005) invesigae he inflaionary effec of financial developmen on economic growh using ime-series daa for Japan over he period They argue ha he relaionship beween financial developmen and economic growh is influenced by he inflaion level in Japan. Specifically, financial developmen has a posiive effec on economic growh when inflaion is lower han 2.5%, and has a negaive effec on economic growh when inflaion is greaer han 2.5%. They explain ha high inflaion means he price level increase and he value of money decrease; consequenly, people would prefer o hold real asses raher han moneary asses as moneary asses value may be affeced negaively by inflaion. However, wihou enough moneary asses, he financial sysem may be suppressed, and hus may be unable o have a posiive influence on economic growh. Apar from inflaion, Greenwood e al. (203) examine he role of echnological progress in he finance-growh relaionship. Using he ime-series daa for he U.S. over he period , hey find ha echnological improvemen in financial inermediaion can conribue o economic growh. They argue ha echnological progress in he U.S. encourages more financial innovaions (e.g., Collaeralized Deb Obligaion and Credi Defaul Swap) and hus he financial capial sysem can be more efficien and effecive in allocaing credi and 0

5 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 capial. Based on heir daa, 29% of GDP is conribued by he echnological improvemens in he financial secor over he period Adusei (203) repors ha financial liberalizaion has a negaive relaionship beween financial developmen and economic growh in Ghana by analyzing annual daa during is found. Adusei (203) explains ha due o lax supervision of he Ghanaian financial sysem, i grans more auonomy o bankers and banking insiuions. By conras, banks are unable o disinguish beween good and bad invesmen projecs due o lack of skilled professionals. Therefore, financial liberalizaion in Ghana has led o over-lending or careless-lending, which has impaced negaively on economic growh. Adusei (203) suggess ha he Ghanaian governmen should ake igher regulaion on he banking secor, especially in heir lending services. Differen from oher counries, China has experienced high economic growh (GDP) over he pas decades wih an average annual growh rae of abou 0%. However, China s financial sysem remains under developed and mainly dominaed by sae-owned banks (Zhang e al., 202). Therefore, he financegrowh relaionship in China aracs grea ineres as a counry-specific sudy. Using 29 provinces daa during , Zheng and Yu (2009) find ha here is a posiive relaionship beween financial developmen and economic growh in China. They sae ha he beer-developed financial sysem can mobilize more savings o invesmen, and hus simulaes economic growh. Wen (2009) finds ha financial developmen has a posiive influence on economic growh using daa from he cenral region of China over he period 978 o Wen (2009) argued ha financial developmen of he cenral region is significan for economic growh and suggesed ha he Chinese governmen should srenghen he reform of financial inermediaion o simulae economic growh. In addiion, Zhang e al. (202) find ha here is a posiive finance-growh relaionship using ciylevel daase of 286 ciies from 200 o As China has enered he World Trade Organizaion (WTO) in 200, many foreign banks were able o provide services in China. Therefore, he domesic banks have o improve heir efficiency in order o become more compeiive han foreign banks. Consequenly, he efficien banking sysem simulaes economic growh by providing more efficien invesmen. By conras, Hasan e al. (2009) find ha here is a negaive relaionship beween financial developmen and economic growh using provincial level daa over he period Tha is because he banking secor was mainly dominaed by sae-owned banks during ha period, and mos of hem had high non-performing loans as hey mainly provided loans for sae-owned enerprise regardless of wheher he enerprise can repay loans. Furhermore, using he provincial level daa for China over , Fang and Jiang (204) examine he effec of financial developmen on primary, secondary, and eriary indusries in China. They find ha financial developmen has no effec on he growh of primary indusry, bu has a posiive effec on he growh of secondary and eriary indusries. Fang and Jiang (204) argue ha he banking secor is mainly making loans o secondary and eriary indusries as he primary indusry has enjoyed slow growh since 998. In his paper, a counry-specific invesigaion is underaken. Our main aim is o analyze he financegrowh relaionship in China using up-o-dae daases. Moreover, he effec of financial developmen on he growh of he primary, secondary, and eriary indusries are examined afer China adoped economic reform in 978. Differen from Fang and Jiang s (204) sudy, his paper uses counry-level daa o examine he finance-growh relaionship and covers a much longer ime period, ha is from 979 o Research mehodology 2.. Sample and variables. We use annual daa from he Naional Bureau of Saisics of China (NBSC) and he World Bank. Specifically, he economic growh and growh of primary, secondary, and eriary indusries, capial sock growh, labor force growh, financial developmen, expor growh rae, and inflaion rae are colleced from above sources from 979 o 203. Dependen variables: as shown in Table, and following he previous lieraure (e.g., King and Levine, 993; Al-Yousif, 2002; Handa and Khan, 2008), his paper uses an annual growh rae of GDP as he proxy of economic growh. The growh rae of GDP is defined as follows: GDP GDP GDP = GDP 00%, where GDP refers o he GDP growh rae in year ; GDP refers o GDP in year ; and GDP - refers o he GDP in he year -. Noably, GDP is measured in percenage. Moreover, and following Fang and Jiang (204), he growh of primary, secondary, and eriary indusry is respecively measured by he annual growh rae of oupu of primary, secondary, and eriary indusry. The primary indusry is a secor of he economy ha makes a direc use of naural resources; he secondary indusry is a secor o produce finished goods by manufacuring he oupus of primary indusry; and he eriary indusry

6 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 is a secor ha provides services o consumers (Clark, 940). In his paper, he growh rae of primary, secondary, and eriary indusry are calculaed as follows: oupu of PI oupu of PI PI = 00%, oupu of PI SI TI oupu of SI oupu of SI = oupu of SI oupu of TI oupu of TI = oupu of TI 00%, 00%, where PI, SI, and TI refer o he oupu growh rae of primary, secondary, and eriary indusry in year, respecively; oupu of PI, oupu of SI and oupu of TI refers o oupus of primary, secondary, and eriary indusry in he year, respecively; oupu of PI -, oupu of SI - and oupu of TI - refers o oupus of primary, secondary, and eriary indusry in year -, respecively. Noably, PI, SI, and TI are measured in percenage. Independen variables: as shown in Table five predicor variables are used in his sudy. Labor force growh is measured by using an annual growh rae of oal populaion. Following Lee and Wong (2005), he oal populaion is used o measure he labor force. The growh rae of he labor force is defined as follows: L P P = P 00%, where L refers o he annual growh rae of he labor force in year ; P refers o oal populaion in year ; and P - refers o oal populaion in year -. Capial growh is measured by using an annual growh rae of gross fixed capial formaion. The gross capial formaion is used o measure how much capial has been formed by an economy during he specific period (NBSC, 203). The capial growh is defined as follows: K K K = K 00%, where, K refers o he capial gross in year ; K refers o gross capial formaion in year ; and K - refers o gross capial formaion in year -. Financial developmen is measured by using an annual growh rae of M2/GDP. McKinnon (973) argue ha financial developmen, ha is developed financial sysem, leads o an increase in using money o make he ransacion in he economy (i.e. no barer). Tha process is called moneizaion of he economy, which is one of he mos imporan indicaors for financial developmen. McKinnon (973) suggesed ha he moneizaion of he economy could be measured by he raio of board money o GDP; he board money, ha is M2, is used o measure he money supply in an economy, and GDP is used o measure he oal oupus of he economy. Therefore, he raio of M2 o GDP could be used o measure he degree of ransacion ha is made by money as a medium of paymen. The financial developmen is defined as follows: ( M 2 / GDP) ( M 2 / GDP) FD = 00%, ( M 2 / GDP) where, FD refers o financial developmen index in year ; (M2/GDP) refers o he level of financial developmen in year ; and (M2/GDP) - refers o he level of financial developmen in year -. The inflaion rae is measured by using annual growh rae of he Consumer Price Index (CPI). We use an annual growh rae of CPI as he proxy of inflaion rae. The inflaion rae is defined as follows: CPI CPI P = 00%, CPI where P refers o he inflaion rae in he year ; CPI refers o he consumer price index in year ; CPI - refers o he consumer price index in year -. Expor growh is measured by using an annual growh rae of expor of goods. The expor of goods refers o he value of commodiies ha are expored across he Chinese border such as expors hrough foreign rade and he gifs and supplies provided by China o oher counries as he aid. According o NBSC (203) China calculaes he value of expors by using Free on Board (FOB) value of goods. FOB value demonsraes he seller is only responsible for he period before he loading of goods, and is no responsible for he furher damage during ranspor. Therefore, he FOB value is he acual value of goods a he momen when i is loaded. The expor growh rae is defined as follows: EX EX EX = 00%, EX where EX refers o expor growh rae in he year ; EX refers o expor value of goods in year ; EX - refers o expor value of goods in year -. 2

7 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 Table. Lis of variables used in he sudy Variables Measuremen Sources Dependen variables Economic growh ( GDP ) annual growh rae of GDP NBSC (203) Growh of primary indusry ( PI ) annual growh rae of oupus in primary indusry NBSC (203) Growh of secondary indusry ( SI ) annual growh rae of oupus in secondary indusry NBSC (203) Growh of eriary indusry ( TI ) annual growh rae of oupus in eriary indusry NBSC (203) Independen variables Labour force ( L ) annual growh rae of oal populaion NBSC (203) Capial ( K ) annual growh rae of gross capial formaion NBSC (203) Inflaion ( P ) annual growh rae of CPI The World Bank (203c) Expor ( EX ) annual growh rae of expor of goods NBSC (203) Financial developmen ( FD ) annual growh rae of M2/GDP NBSC (203) Noe: sample period is over he period Proposed models. Following Lee and Wong (2005), a muliple regression model is used o analyze he relaionship beween financial developmen and economic growh as shown below: GDP = α+ βl + β2k + β3fd + () + β EX + β P + e 4 5, where GDP refers o economic growh in year ; L refers o labor force growh in year ; K refers o capial growh in year ; FD refers o financial developmen index in year ; EX refers o expor growh rae in year ; P refers o he inflaion rae in year ; α is he inercep of regression line; β are he marginal effec of independen variables; e is he error erm. Moreover, similar regression models are se o examine he effec of financial developmen on he growh of primary, secondary, and eriary indusry in China, which is shown as follows: PI = α + βl + β 2K + β3fd + β 4EX + (2) + β P + e, 5 Table 2. Descripive saisics SI = α + β L + β K + β FD + β EX + + β P e, TI = α+ β L + β K + β FD + β EX + + β P e, (3) (4) where P I, S I, T I denoes he oupu growh rae of primary, secondary, and eriary indusry in year, respecively; and he remaining symbols have same definiion as in equaion () above. 3. Resuls 3.. Descripive saisics. Table 2 shows summary saisics of all variables. Noably, he uni of hem is percenage as hey are measured in growh rae. The minimum and maximum values of annual GDP growh raes G DP are 3.80% and 5.2%, respecively; and GDP has grown a an average annual rae of 9.8% during Due o he liberalizaion of foreign rade and invesmen, expors and capial have experienced a high growh rae afer 978. The expor growh rae ( ) and capial growh rae (K ) have averaged 22.45% and 6.84%, respecively. Variables Mean Sd. Dev. Min Max Skewness Kurosis G DP P I S I T I L K P

8 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 Naughon (2007) argue ha he 978 markeoriened reform can lead o changes in indusrial srucure. Specifically, China has ransferred is aenions from primary indusry o secondary and eriary indusry afer 978. According o he descripive saisics shown in Table 2, he growh rae of primary indusry ( P I ) has averaged 4.58% during Differen from primary indusry, he secondary and eriary indusries have much higher average annual growh rae, and boh of hem have more han wice he growh rae of primary indusry. Specifically, he annual growh rae of secondary indusry ( S I ) has averaged.29% and he annual growh rae of eriary indusry ( T I ) has averaged 0.78%. The populaion (L) has grown a an average rae of 0.99% and he growh rae of M2/GDP ( ) has averaged 5.5 during The inflaion rae (P ) has averaged 5.39%, which is higher han he inflaion rae (averaged 3.8%) in Japan, bu similar o he inflaion rae (averaged 5.%) in Taiwan as repored by Lee and Wong (2005). Dielman (200) saes ha mulicollineariy can decrease he accuracy of he esimaed regression model as independen variables are highly correlaed wih each oher, and hus he real effec of an individual independen variable on dependen variable can be disored by oher independen variables. According o Dielman (200), if he Table 3. Correlaion coefficiens absolue value of he correlaion coefficien beween independen variables is smaller han 0.80, he mulicollineariy problem can be rejeced. As shown in Table 3, Panel A presens he correlaion beween various independen variables, and PaneB presens he correlaion beween various indusries. The correlaion beween inflaion rae (P ) and capial growh rae (K ) has he highes value of 0.534; and he correlaion beween financial developmen ( ) and capial growh (K ) has he lowes value of Tha is, he correlaion coefficiens beween all independen variables range from o 0.534, wih heir absolue values being lower han 0.80, and hus suggesing ha no serious mulicollineariy problems exis. Besides he correlaion coefficien, he mulicollineariy problem is esed by compuing olerance coefficien and Variance Inflaion Facor (VIF) of each independen variable (see more deails in Panel C of Table 3). According o Dielman (200), if he olerance value is greaer han 0. and VIF is smaller han 0, he rejecion of he mulicollineariy problem can be acceped. Based on our VIF resuls he value of he olerance coefficien of all independen variables is ranging from 0.44 o 0.63, which is greaer han 0.; and he VIF of all independen variables is ranging from.63 o 2.43, which is smaller han 0. Therefore, he rejecion of mulicollineariy problem can be acceped, and all five independen variables can be used in our proposed regression models. Panel A L K P Panel B PI L K P.28 * (.05) *** (.007) (.493) *.62 (.39) (.044) (.77) -.38 **.405 ***.534 ***.40 *** (.02) (.008) (.000) (.008) PI SI TI SI TI Panel C. Collineariy es -.2. (.223) ** (.053) (.00) Independen variables Tolerance coefficien VIF L

9 Panel C. Collineariy es Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 Table 3 (con.). Correlaion coefficiens Independen variables Tolerance coefficien VIF K P Noe: ***, **, and * denoes significance level of correlaion is 0.0, 0.05, 0., respecively Empirical resuls. Table 4 shows he resuls for he economic growh ( G DP ) regression model. The resuls sugges ha financial developmen has a negaive effec on economic growh a he 99% confidence level, and he marginal effec of financial Table 4. Regression model for Coefficien Sandard error -saisic p-value developmen on economic growh is Tha is, increasing % of he financial developmen index ends o decrease he annual GDP growh rae by 0.67% when oher independen variables, such as inflaion and labor raes are aken ino accoun. G DP Confidence Inerval Lower 95% Upper 95% Inercep L K P Model parameers R R 2 adj F-saisic P-value Noe: G DP is he dependen variable. Our resul is consisen wih oher empirical sudies, such as Hasan e al. (2009) and Adusei (203) who find ha here is a negaive relaionship beween financial developmen and economic growh in China and Ghana. Hasan e al. (2009) argue ha he negaive finance-growh relaionship in China is caused by high non-performing loan in he saesowned banks. They poin ou ha he Chinese banking secor is mainly dominaed by sae-owned banks. However, many sae-owned banks do no perform well as hey mainly provide loans o saeowned enerprise regardless of wheher he enerprise can repay loans. Moreover, Adusei (203) argues ha he negaive finance-growh relaionship in Ghana is caused by he lax supervision of he governmen, which leads o overlending and careless-lending in banking secors. Our resuls also show ha he annual growh rae of CPI (P ) is negaively associaed wih economic growh a he 99% confidence level. The annual growh rae of gross capial formaion (K ) is posiively associaed wih he economic growh a he 99% confidence level. More han 65% of he variaion in he economic growh ( G DP) is explained by hree predicors, namely he annual growh rae of gross capial formaion (K ), he financial developmen ( ) and he annual growh rae of CPI (P ), as shown in Table 4. By conras, some empirical sudies show a posiive relaionship beween economic growh and financial developmen in China (see for example, Zheng and Yu, 2009; Zhang e al., 202). We sugges ha hese conflicing resuls wih our findings are mainly due o he ime-period selecion. Specifically, he ime period seleced by Zheng and Yu (2009) and Zhang e al. (202) does no include he firs phase of economic reform (i.e. before 994). However, our paper s ime-frame (i.e ) includes boh he firs phase (i.e ) and he second phase (i.e. 994 o dae) of economic reforms ha have been pursued in China. Table 5 shows he resuls for he growh in primary indusry ( P I ) regression model. Since he P-value of he coefficien of he financial developmen is greaer han 0., herefore, financial developmen 5

10 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 has no significan effec on he growh of primary indusry. The resul is consisen wih Fang and Jiang (204) findings who also found ou ha financial developmen has no significan influence on he growh of he primary indusry. Fang and Jiang (204) explain ha he banking secor is mainly making loans o secondary and eriary indusries, as he primary indusry has enjoyed slow growh since 998. Our resuls also show ha he primary indusry is less developed han he secondary and eriary indusry over he period Specifically, he annual growh rae of secondary and eriary indusry has averaged Table 5. Regression model for Coefficien Sandard error -saisic p-value.29% and 0.78%, respecively, which is more han wice he average annual growh rae of he primary indusry (4.58%). The developmen rends of he hree indusries in China are consisen wih he hree-secor heory carried ou by Clark (940). Clark (940) argues ha counries end o shif heir aenion from primary indusry (i.e. secor for proving goods) o secondary indusry (i.e. secor for providing goods) and from he secondary indusry o eriary indusry (i.e. secor for providing services) along wih he economic developmen as he demand for goods and services will increase along wih he economic developmen. P I Confidence Inerval Lower 95% Upper 95% Inercep L K P Model parameers R R 2 adj F-saisic 0.97 P-value Noe: P I is he dependen variable. Table 6 shows he resuls for he growh in secondary indusry ( S I ) regression model. The P-value of coefficien of financial developmen ( ) is also greaer han 0.0, which means ha financial developmen has no significan effec on he growh of secondary indusry. However, our resuls show ha he annual growh rae of gross Table 6. Regression model for S I Coefficien Sandard error -saisic p-value capial formaion ( K ) is posiively associaed wih he growh of he secondary indusry ( S I ) a he 99% confidence level. More han 60% of he variaion in he secondary indusry ( S I ) is explained by only one variable namely he annual growh rae of gross capial formaion ( K ), as shown in Table 6. Confidence Inerval Lower 95% Upper 95% Inercep L K P Model parameers R R 2 adj F-saisic 0.34 P-value Noe: S I is he dependen variable. 6

11 Table 7 shows ha financial developmen has a negaive effec on growh of eriary indusry a he 99% confidence level. The marginal effec of financial developmen on he growh of eriary indusry is Tha is, increasing % of he financial developmen index ends o decrease he oupu growh rae of eriary indusry by 0.332% when oher independen variables remain consan. In line wih his finding, he annual growh rae of CPI ( P ) is also Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 Table 7. Regression model for T I Coefficien Sandard error -saisic p-value negaively associaed wih economic growh a he 99% confidence level. By conras, he annual growh rae of gross capial formaion (K ) and he annual growh rae of oal populaion (L ) are boh posiively associaed wih economic growh a he 99% confidence level. Finally more han 60% of he variaion in he eriary indusry ( T I ) is explained by he four significan variables as shown in Table 7. Confidence Inerval Lower 95% Upper 95% Inercep L K P Model parameers R R 2 adj F-saisic P-value Noe: T I is he dependen variable. Our resuls are in conflic wih hose by Fang and Jiang (204). Their empirical resuls showed ha here is a posiive associaion beween financial developmen and he growh of boh secondary and eriary indusries. We argue ha here can be hree main reasons o jusify hose differen resuls. Firsly, we use differen ime-frame (i.e., ), bu similar o Zheng and Yu (2009) and Zhang e al. (202) sudies, while Fang and Jiang (204) chose ime-period ha only covers he second phases of economic reform. However, our invesigaion sample covers boh firs and second phase of economic reform. As explained above, in he firs phase of economic growh, he financial developmen ends o impede economic growh, and in he second phase of economic reform, financial developmen ends o spur he economic growh. Therefore, i is no surprising ha Fang and Jiang (204) found ou he posiive relaionship beween financial developmen and growh in secondary and eriary indusry. Secondly, our indicaor of financial developmen is differen. Paricularly, Fang and Jiang (204) used he raio of bank loan o GDP as he proxy of financial developmen, whils we use he raio of M2 o GDP as he proxy for financial developmen. Finally, we use a differen se of explanaory variables, as shown in Table. Fang and Jiang (204) used differen se of predicor variables apar from only one variable namely labor force when hey examine he relaionship beween financial developmen and growh in each indusry. In our sudy we use five explanaory variables o examine he relaionship beween financial developmen and growh in each indusry following he framework of Lee and Wong (2005). In regression analysis, if some of he explanaory variables are omied, he resuls migh be differen (Koop, 2000). We argue ha his migh explain he difference in resuls beween our sudy and Fang and Jiang (204). Conclusions Differen from he mos previous sudies in China, his paper uses four muliple regression models o examine he effec of financial developmen on economic growh afer 978; and o examine he effec of financial developmen on he growh of he primary, secondary, and eriary indusries. Five independen variables, namely financial developmen, labor force, capial growh, expor growh and inflaion rae, are used in his invesigaion. Our empirical resuls show ha financial developmen has a negaive effec on economic growh in general bu he growh of he eriary indusry in paricular. By conras, our findings indicae ha financial developmen has no effec on he growh of boh he primary and secondary indusries. We argue ha he high raio of M2/GDP in China can be affeced by he economic reform and he fac ha he Chinese capial marke is underdeveloped. We recommend ha he Chinese 7

12 Invesmen Managemen and Financial Innovaions, Volume 2, Issue 3, 205 governmen may need o pay more aenion on developing he capial marke, and hus provide more opions for Chinese households o disribue heir money. In addiion, he governmen may provide more opions for enerprises o obain finance for invesmen. Fuure research could consider alernaive measures for financial developmen according o China s References specific condiions. As he annual growh rae of oal populaion is used in our paper o measure labor force, fuure research could be exended o by using working-age populaion o invesigae wheher differen resuls could be found. Finally, o exend he analysis o invesigae he causaliy effecs beween financial developmen and economic growh.. Adusei, M. (203). Financial developmen and economic growh: Evidence from Ghana, The Inernaional Journal of Business and Finance Research, 7 (5), pp Al-Yousif, Y.K. (2002). Financial developmen and economic growh anoher look a he evidence from developing counries, Review of Financial Economics, (2), pp Akimov, A., Wijeweera, A., and Dollery, B. (2009). Financial developmen and economic growh: Evidence from ransiion economies, Applied Financial Economics, 9 (2), pp Akinboade, O.A. (2000). The relaionship beween financial deepening and economic growh in Tanzania, Journal of Inernaional Developmen, 2 (7), pp Apergis, N., Filippidis, I., and Economidou, C. (2007). Financial deepening and economic growh linkages: A panel daa analysis, Review of World Economics, 43 (), pp Arcand, J.L., Berkes, E., and Panizza, U. (202). Too Much Finance? Inernaional Moneary Fund. 7. Čihák, M., Demirgüç-Kun, A., Feyen, E., and Levine, R. (202). Benchmarking Financial Sysems around he World, Working paper (No 675), World Bank, Washingon. 8. Clark, C.G. (940). Condiion of economic progress. London: Macmillan Press. 9. Dielman, T.E. (200). Applied regression analysis for business and economics. Pacific Grove, CA: Duxbury/Thomson Learning. 0. Fang, X. and Jiang, Y. (204). The promoing effec of financial developmen on economic growh: Evidence from China, Emerging Markes Finance and Trade, Vol. 50, pp Ghali, K.H. (999). Financial developmen and economic growh: The Tunisian experience, Review of Developmen Economics, 3 (3), pp Greenwood, J., Sanchez, J.M. and Wang, C. (203). Quanifying he impac of financial developmen on economic developmen, Review of Economic Dynamics, 6 (), pp Gregorio, J. and Guidoi, P.E. (995). Financial developmen and economic growh, World Developmen, 23 (3), pp Handa, J. and Khan, S.R. (2008). Financial developmen and economic growh: A symbioic relaionship, Applied Financial Economics, 8 (3), pp Hasan, I., Wachel, P. and Zhou, M. (2009). Insiuional developmen, financial deepening and economic growh: Evidence from China, Journal of Banking & Finance, 33 (), pp King, R.G., and Levine, R. (993). Finance and growh: Schumpeer migh be righ, The Quarerly Journal of Economics, Vol. 08 No. 3, pp Law, S.H., and Singh, N. (204). Does oo much finance harm economic growh? Journal of Banking and Finance, Vol. 4, pp Lee, C. and Wong, S.Y. (2005). Inflaionary hreshold effecs in he relaionship beween financial developmen and economic growh: evidence from Taiwan and Japan, Journal of Economic Developmen, 30 (), p McKinnon, R. (973). Money and Capial in Economic Developmen. Washingon, DC: Brookings Insiuion. 20. Naional Bureau of Saisics of China (NBSC) (203). China Saisical Yearbook Naughon, B. (2007). The Chinese economy: Transiions and growh, MIT press. 22. Qin, D. (2003). Deerminans of household savings in China and heir role in quasi-money supply, The Economics of Transiion, (3), pp Schumpeer, J.A. (9). The Theory of Economic Developmen: An Inquiry ino Profis, Capial, Credi, Ineres, and he Business Cycle, Cambridge, MA: Harvard Universiy Press. 24. Shaw, E.S. (973). Financial Deepening in Economic Developmen, New York: Oxford Universiy Press. 25. Wen, Y. (2009). Financial Developmen and Economic Growh in Cenral Region of China: An Empirical Analysis, In Business Inelligence and Financial Engineering, Inernaional Conference on, pp Yueh, L. (203). Wha drives china s growh? Naional Insiue Economic Review, 223 (), pp Zhang, J., Wang, L. and Wang, S. (202). Financial developmen and economic growh: Recen evidence from china, Journal of Comparaive Economics, 40 (3), pp Zheng, C. and Yu, Y. (2009). Financial Developmen and Economic Growh Based on he Panel Daa ( ) of All Provinces in China, In Business Inelligence and Financial Engineering, Inernaional Conference on, pp