Short-Term International Capital Flows: The Case of China

Size: px

Start display at page:

Download "Short-Term International Capital Flows: The Case of China"

Prosper Clarke
5 years ago
Views:

1 PROSIDING PERKEM VII, JILID 1 (2012) ISSN: X Short-Term Iteratioal Capital Flows: The Case of Chia Ta Ju Ju tajj@mail.ustc.edu.c Masor Jusoh Tamat Sarmidi School of Ecoomics Faculty of Ecoomics ad Maagemet The Natioal Uiversity of Malaysia ABSTRACT This paper tries to ivestigate dyamic relatioships betwee iteratioal short-term capital flows ad Chia s ecoomy which icludig iterest rate, exchage rate, stock price ad real estate price from 1999Q1 to 2011Q4. We use Augmeted Dickey Fuller test (ADF) to test the statioary of time series data, the make the Johase Coitegratio test ad apply Grager Causality test to aalyze the relatioships amog the variables. The study fids that the foreig capital flows impacted by the foreig exchage rate fluctuatios ad stock price movemets rather tha iterest rate differetial; ad the cotiued rise of housig price may partially explaied by the iflow of huge foreig capital ito Chia. Keyword: Iteratioal capital flows, iterest rate, exchage rate, real estate price, stock price, Grager Causality test INTRODUCTION Alog with the era of ecoomic globalizatio ad fiacial liberalizatio, the iteratioal capital is takig o a icreasigly acceleratig tred ad playig a more promiet role i today s world ecoomy. Similarly, as the world's secod largest ecoomy after the Uited States, Chia is the world's fastest-growig major ecoomy with growth rates averagig 10% over the past 30 years sice the implemetatio of reform ad opeig up policy. Chia's high icreasig ecoomy is to chage the charm attracts global ivestors eyes. The foreig capital etered ito Chia's idustry ad commerce extesively, while brigs the ecoomic beefits to us, it also browbeats the Chia's fiacial market move smoothly i the past decades. Sice the early 1990s, the govermet has allowed foreig ivestors to established joit vetures, provided some assuraces agaist atioalizatio, allowed foreig parters to become chairs of joit veture boards, ad authorized the establishmet of wholly foreig-owed eterprises, ow the preferred form of Foreig Direct Ivestmet (FDI). Ad the foreig ivestmet remais a strog elemet i Chia's rapid expasio i world trade ad has bee a importat factor i the growth of job opportuities. Over 480 out of Fortue 500 eterprises have already ivested i Chia. I 2007 aloe, Chia approved 37,888 ew foreig busiess establishmets. The Chiese govermet has begu relaxig of more tha a decade of tight capital cotrols for foreig ivestors to put moey ito Chia s stock market ad other fiacial ivestmets. To be more specific, Chia s foreig exchage regulator (the State Admiistratio of Foreig Exchage, SAFE) is to loose its cotrol over the Qualified Foreig Istitutioal Ivestor (QFII) program, which allows overseas istitutios to ivest i Chiese securities market. Ad Chia also authorized foreig fiacial idustries brig their busiess to Chia ad allowed foreig ivestors to purchase special B shares of stock i selected compaies listed o the Shaghai ad Shezhe Securities Exchages. As of the ed of March 2011, the foreig exchage reserves of Chia holds $ trillio, makig it the highest foreig exchage reserve i the world ad far exceeded holdigs of the ext largest holder, Japa (about $1 trillio). But Chia s trade surplus ad FDI partially explai this, deductig ivestmet icome ad reserves still leaves a uexplaied residual of $214 billio. Some ecoomists use this as a proxy for hot-moey iflows. But some of it may reflect o-speculative trasactios, such as foreig borrowig by Chiese firms. Accordig to the report, it estimates that Chia received up to 168 billio U.S. dollars i hot moey i This far exceeds aythig previously experieced by ay emergig ecoomy. I recet years, with the appreciatio of RMB strogly expected, large foreig speculative fuds have poured ito Chia supposedly seekig short-term profits, especially i Chiese estimates of the amout of iteratioal speculative capital i Chia vary from $500 billio to $1.75 trillio, severely affectig the Chia s ecoomic ad fiacial security, as well as have egative Persidaga Kebagsaa Ekoomi Malaysia ke VII (PERKEM VII) Trasformasi Ekoomi da Sosial Ke Arah Negara Maju Ipoh, Perak, 4 6 Ju 2012

2 Prosidig Persidaga Kebagsaa Ekoomi Malaysia Ke VII cosequeces for the U.S. ad global ecoomy. The iflux of iteratioal speculative capital is cotributig to Chia s already existig problems with iflatio. While speculate i the appreciatio of Chiese RMB, the short-term iteratioal capital also hut for arbitrage opportuities i securities markets ad real estate markets. The latest statistics suggest that, the securities markets ad real estate markets has boomed sice the RMB exchage rate mechaism reform i 2005.To be more specific, real estate prices soar ad up to a average rise of 12% eve doublig i some large cities like Beijig, Shaghai ad Shezhe, high profits also lured the a large iflow of foreig capital ito Chia every year. The mai research objective of this study ivolves usig the ecoometric model which primarily aims to examie how short-term iteratioal capital flows affects Chia s macro ecoomy, we applied the quarterly data of iterest rate differetial, omial exchage rate, exchage rate expectatio, stock price ad real estate price of Chia from 1999 to 2011 as idicators to aalyzed the ifluece of ecoomy brought by short-term iteratioal capital flows. RELEVANT LITERATURE REVIEW I geeral, the academy classified the iteratioal log-term ad short-term capital by the period for repaymet. Short-term iteratioal capital meas the duratio of the iteratioal ivestmets or loa withi oe year, it cotais marketable securities (short-term ivestmets), short-term loas, bak Istrumets, foreig exchage bill discout ad trade fiacig etc. Both Obstfeld (1994) ad Kevi (2000) believed that capital flows could spread the ivestmet risks, promote specializatio of productio ad reasoable distributio of capital i the markets of the developig coutries. But a few experts offer differet view ad proposal, Alejadro Lopez-Mejia (1999) poited out that iteratioal capital iflow i big scale is possible to cause moey creatio, exchage depreciatio ad iflatio, so the iteratioal short-term capital flows would probably impede ecoomic developmet. Bosworth ad Collis (1999) researched 20 developig coutries with capital iflow by usig multi-variace aalysis ad obtaied that foreig direct ivestmet (FDI) has a strog ifluece o the ivestmet of capital iflow coutries, the stadard regressio coefficiet of them was approaches Ediso ad Levie (2002) clarified the relatioship betwee Iteratioal Fiacial Itegratio ad ecoomic growth based o the way of ecoometrics metrology, ad they got that there does ot exist explicit relatioship betwee those two. She Kurog (1998) aalyzed the correlatio betwee iteratioal capital iflow ad ecoomic growth of Chia, ad he drew a coclusio that Chia is still a coutry with capital shortage from geerally speakig. I order to maitai ecoomic growth, the oly effective way is to rely o foreig ivestmets to support. He also offered some costructive advices for attractig foreig capital such as improvig the eviromet for ivestmet, adjustig idustry structure ad icreasig policy supports from govermet. Wag Xi (2003) carried out regressig aalyses o the foreig exchage settlemet ad sales busiess of Chia from Jue 1999 to May 2002 by usig least square method, ad the empirical result demostrated that the capital iflow was greatly iflueced by short-term assets yield i this period. Li Zeguag, Fufei ad Tag Weixia (2003) explored by measurig the capital volatility coefficiets ad foud that foreig loas ad portfolio ivestmet are more sesitive to the volatility of iteratioal short-term capital flows tha FDI, so the effects of those two factors are most sigificat to Chia s real ecoomy. Wag Qi (2006) proposed a ifluecig factors model for Chia s iteratioal capital flows accordig to iteratioal stadard, ad he isisted that Chia s capital movemets are maily affected by iterest rate, foreig exchage rate, iflatio rate, ecoomic opeess, atioal policy ad so o through multiple regressio aalysis. Sog Bo ad Gao Bo (2007) set up Error Correctio Model (ECM) by usig the quarterly data of disbursemet of foreig capital ad real estate price from 1999 to 2006 i cosideratio of a eviromet of cotiued iflatio. Ad Grager Causal Relatio Test idicated that jack-up of housig price results i iflow of foreig capital i the short ru; while i the log ru, iflow of foreig capital has a effect o housig price shot up. So they thought it is advatageous for Chia to keep stabilizatio of housig price by cotrollig excessive iflow of foreig capital i curret circumstace. Zhag Yihao (2007) set up a arbitrage model of iterest rate, foreig exchage rate ad price; he foud that the factors related to iterest rate, foreig exchage rate ad price are also have strog effects to iteratioal short-term capital movemets.

3 74 Ta Ju Ju, Masor Jusoh, Tamat Sarmidi MODEL AND DATA The Mudell-Flemig model is a ecoomic model first set forth by Robert Mudell ad Marcus Flemig. The model is a extesio of the IS-LM model. Whereas the traditioal IS-LM Model deals with ecoomy uder autarky (or a closed ecoomy); the Mudell-Flemig model tries to describe a ope ecoomy. Typically, the Mudell-Flemig model portrays the relatioship betwee the omial exchage rate ad a ecoomy's output (ulike the relatioship betwee iterest rate ad the output i the IS-LM model) i the short ru. The Mudell-Flemig model has bee used to argue that a ecoomy caot simultaeously maitai a fixed exchage rate, free capital movemet, ad a idepedet moetary policy. The study follows Che Lagyu & Che Yu (2009), we propose the extesio of Mudell Flemig model (Mudell, Robert A. (1963) & Flemig, J. Marcus (1962)) to explai as follow: (1) Iteratioal short-term capital flows (SCF) For SCF, we follow the previously studies that use the World Bak Residual Method, which was firstly proposed to measure the capital flight i the World Developmet Report 1985 released by World Bak. I World Bak s poit of view, capital flight caot be measured directly, but ca be roughly estimated as a residual. Which assumes the capital flight is take as a residual of four compoets: the Icrease i Official Foreig Reserves (IOFR), the Icrease i Exteral Debt (IEB), Net Foreig Direct Ivestmet (NFDI) ad Curret Accout Deficit (CAD). So the formula of capital flight would be described as follows: SCF = IEB + NFDI + CAD IOFR (2) (2) Iterest Rate Differetial (IRD) The relative chage i iterest rate differetial affects the iteratioal short-term capital flows seriously, ad the capital usually flows from a coutry with low iterest rate to a higher oe. The hypothesis is that the wider spreads the iterest rate betwee two coutries, the larger the iteratioal short-term capital that will iflow, other thigs beig equal. We choose Chia Time Deposit Rate (oe year) as Chia s domestic iterest rate ad US dollar (oe-year iterest rate) as foreig iterest rate. Ad we use proportioality of them to represet the iterest rate differetial. The sources of statistic data are output directly from the software of DataStream. (3) Nomial Exchage Rate (NER) As well, the appreciatio i Chiese RMB also attracts foreig capital iflow for gettig profits. Because of RMB exchage rate reform, this study chooses the omial exchage rate of RMB to US Dollar as idicator. The exchage rate is expressed i uits of domestic currecy price of foreig currecies i direct quotatio system, the umbers upward meas RMB s devaluatio ad vice versa. Data also comes from DataStream. (4) Stock Market's Retur (SMR) This study picks the stock market's retur of Chia to aalyze the ifluece o securities markets brought by iteratioal short-term capital flows. Specifically, we apply the Shaghai Composite Idex (SCI) as the stock price, ad the statistic data from official website of Shaghai Stock Exchage. (5) Real Estate Market's Retur (REMR) For study the impact betwee iteratioal short-term capital flows ad Chia s real estate price, we choose the statistics data of housig price after the real estate market reformed i The data we selected are released by Natioal Bureau of Statistics of Chia. (1)

4 Prosidig Persidaga Kebagsaa Ekoomi Malaysia Ke VII RESEARCH METHODOLOGY (1) The Augmeted Dickey Fuller (ADF) Test I statistics ad ecoometrics, a Augmeted Dickey Fuller test (ADF) is a test for a uit root i a time series sample. It is a augmeted versio of the Dickey Fuller test for a larger ad more complicated set of time series models. The Augmeted Dickey Fuller (ADF) statistic is a egative umber whe used i the test. The more egative it is, the stroger the rejectio of the hypothesis that there is a uit root at some level of cofidece. (2) Johase Coitegratio Test I statistics, the Johase test is a procedure for testig coitegratio of several I(1) time series. This test permits more tha oe coitegratig relatioship so is more geerally applicable tha the Egle Grager test which is based o the Dickey Fuller (or the augmeted) test for uit roots i the residuals from a sigle (estimated) coitegratig relatioship.there are two types of Johase test, either with trace or with eigevalue, ad the ifereces might be a little bit differet. The ull hypothesis for the trace test is the umber of coitegratio vectors r?, the ull hypothesis for the eigevalue test is r =?. (3) Vector Auto Regressio (VAR) Models Vector Auto Regressio (VAR) is a statistical model used to capture the liear iterdepedecies amog multiple time series. VAR models geeralize the uivariate autoregressio (AR) models. All the variables i a VAR are treated symmetrically; each variable has a equatio explaiig its evolutio based o its ow lags ad the lags of all the other variables i the model. A VAR model describes the evolutio of a set of k variables (called edogeous variables) over the same sample period (t = 1... T) as a liear fuctio of oly their past evolutio. The variables are collected i a k x 1 vector, which has th as the i elemet y i, t the time t observatio of variable y i. (4) The Grager Causality Test The Grager causality test is a statistical hypothesis test for determiig whether oe time series is useful i forecastig aother. Clive Grager argued that there is a iterpretatio of a set of tests as revealig somethig about causality. The test ivolves estimatig the followig pair of regressios: Y X Y + u t i t i j t j 1t i1 j1 X Y X + u t i t i j t j 2t i1 j1 where it is assumed that the disturbaces u 1 t ad u 2 t are ucorrelated. If variable X (Grager) causes variable Y, the chages i X should precede chages i Y. Therefore, i a regressio of Y o other variables (icludig its ow past values) if we iclude past or lagged values of X ad it sigificatly improves the predictio of Y, the we ca say that X (Grager) causes Y. A similar defiitio applies if Y (Grager) causes X. EMPIRICAL RESULTS Table 1 reports the results of ADF test. The test cosistetly suggest that all variables icludig ΔSCF, ΔIRD, ΔNER, ΔREMR, ΔSMR are statioary i levels I (1).

5 76 Ta Ju Ju, Masor Jusoh, Tamat Sarmidi TABLE 1: Results of Augmeted Dickey-Fuller Uit Root Test First Differece Variable Test critical values t-statistic Prob.* 1% 5% 10% ΔSCF ΔIRD ΔNER ΔREMR ΔSMR Before take tests, we should idetify the order of lag legth for the urestricted VAR at first. The optimal lag legth is selected accordig to a set of statistical selectio iformatio criterios like Fial predictio error (FPE), Akaike iformatio criterio (AIC), Schwarz criterio (SC) ad Haa-Qui iformatio criterio (HQ). Table 2 provides the appropriate lag legth for the estimated VAR models. The optimal lag legth for the chose VAR models is five lags. TABLE 2: Lag legth selectio Lag LogL LR FPE AIC SC HQ NA 1.15e * * * * * * idicates lag order selected by the criterio LR: sequetial modified LR test statistic (each test at 5% level) FPE: Fial predictio error AIC: Akaike iformatio criterio SC: Schwarz iformatio criterio HQ: Haa-Qui iformatio criterio TABLE 3: The results of Johase Coitegratio Test Urestricted Coitegratio Rak Test (Trace). Hypothesized No. of CE(s) Eigevalue Trace Statistic 0.05 Critical Value Prob.** Noe * At most 1 * At most At most At most Trace test idicates 2 coitegratig eq(s) at the 0.05 level * deotes rejectio of the hypothesis at the 0.05 level **MacKio-Haug-Michelis (1999) p-values Urestricted Coitegratio Rak Test (Maximum Eigevalue) Hypothesized No. of CE(s) Eigevalue Max-Eige Statistic 0.05 Critical Value Prob.** Noe * At most At most At most At most Max-eigevalue test idicates 1 coitegratig eq(s) at the 0.05 level * deotes rejectio of the hypothesis at the 0.05 level ** MacKio-Haug-Michelis (1999) p-values From the results of Johase Coitegratio test, the Trace test ad Maximum Eigevalue test idicated that 2 coitegratig eq(s) ad 1 coitegratig eq(s) separately at the 0.05 level. The empirical results

6 Prosidig Persidaga Kebagsaa Ekoomi Malaysia Ke VII deote that the short-term iteratioal capital flows, iterest rate, exchage rate, real estate markets retur ad stock markets retur have a steady equilibrium relatioship i the log ru. Appedix shows the results of VAR Grager Causality/Block Exogeeity Wald Tests with lag legth of 5. The test results suggest that there is strog evidece of a uidirectioal causal likage that rus from Nomial Exchage Rate (NER) to Short-term Iteratioal Capital Flows (SCF) ad Stock Market Retur (SMR) to SCF. Moreover, the causality fidigs show that SCF rejects the Null Hypothesis, so it does Grager Cause Real Estate Market Retur (REMR). This may explai why the cost of houses have bee risig a estimated 25% aually i Beijig ad Shaghai i recet years. The costat iflux of foreig capital to Chia real estate markets cotributes to the precipitous rise of housig prices to a great extet. CONCLUSION The preset study empirically ivestigates the relatioships amog the Short-term Iteratioal Capital Flows, Iterest Rate Differetial, Nomial Foreig Exchage Rate, Real Estate Markets Retur ad Stock Markets Retur. Ad the empirical results idicated that, the foreig exchage rate ad equity markets retur have strog iflueces to short-term iteratioal capital flows. I sum, the appreciatio of Chiese RMB ad the potetial value of the stock would attract tos of short-term iteratioal capital iflow ito Chia s capital markets. Especially the call for a appreciatio of the Chiese RMB heated up, ad it would directly affect the retur rate betwee RMB ad other foreig exchage assets. With the appreciatio of RMB strogly expected, large foreig speculative capital has poured ito real estate idustries ad fiacial markets of Chia, severely affectig the state s ecoomy ad fiacial security. The results also show that the Iterest rate is ot the critical factor determiig the flows of iteratioal short-term capital. Eve low iterest rate differetial exists i differet capital markets, the ivestors are eager to get a higher profit rather tha the hope of cashig i o the Chiese currecy's steady rise i value. The fidigs also provide robust evidet that the high real estate price of Chia largely o the stregth of support from short-term iteratioal capital flows. So the iflux of iteratioal speculative capital is cotributig to Chia s already existig problems with iflatio. REFERENCES Alejadro LM. (1999). Large capital Flows: A Survey of The Causes, Cosequeces, ad Policy Resposes [J]. IMF Workig Papers. 1999(3): Barry Bosworth & Susa Collis. (1999). Capital Flows to Developig Ecoomies: Implicatios for Savig ad Ivestmet, Brookigs Papers o Ecoomic Activity: Che Lagyu & Che Yu. (2009). Exchage Rate of RMB, Price of Property ad Short-term Iteratioal Capital Flows. Ecoomics Maagemet. 2009(1): 1-6. David A. Dickey & Waye A. (1979). Fuller. Distributio of the Estimators for Autoregressive Time Series with a Uit Root. Joural of the America Statistical Associatio, Vol. 74, No. 366: Ediso, Hali J. & Levie, Ross & Ricci, Luca & Slok, Torste, (2002). "Iteratioal fiacial itegratio ad ecoomic growth," Joural of Iteratioal Moey ad Fiace, Elsevier, vol. 21(6): Gujarati. (2004). Basic Ecoomics, Fourth Editio. The McGraw-Hill Compaies: J. Marcus Flemig. (1962). Domestic Fiacial Policies uder Fixed ad uder Floatig Exchage Rates. Iteratioal Moetary Fud, Vol. 9, No. 3 (Nov. 1962): Li Zeguag, Fu Fei & Tag Weixia. (2003). The Aalysis of Volatility of Capital ad Its Ecoomic Effects. Joural of Fiace ad Ecoomics. 2003(12): Obstfeld. (1994). Risk-takig, Global Diversificatio, ad Growth [J]. America Ecoomic Review. (5): Robert F. Egle & C. Grager. (1987). Co-Itegratio ad Error Correctio: Represetatio, Estimatio, ad Testig. Ecoometrica, Vol. 55, No. 2: She Kurog. (1998). Short-term Iteratioal Capital Iflow ad Chia s Ecoomic Growth. Chia Idustrial ecoomy. 1998(10): Sog Bo & Gao Bo. (2007). The Impact of Iteratioal Capital Flows o Real Estate Markets. Fiacial ad Ecoomic Questios Research. 2007(3):

7 78 Ta Ju Ju, Masor Jusoh, Tamat Sarmidi Wag Qi. (2006). The Model Costructio ad Aalysis of Chia s iteratioal Capital Flows. Iteratioal Fiace Research. 2006(6): Wag Xi. (2003). The Relative Factors of Short-term Capital Iflow ito Chia. Iteratioal Fiace Research. 2003(1): World Bak. (1985). World Developmet Report Oxford Uiversity Press: Page No.64. Zhag Yihao, Pei Pig & Fag Xiamig. (2007). The Motivatio of Short-term Iteratioal Capital Iflow ito Chia. Iteratioal Fiace Research. 2007(9): APPENDIX VAR Grager Causality/Block Exogeeity Wald Tests Depedet variable: SCF IRD NER REMR SMR All Depedet variable: IRD SCF NER REMR SMR All Depedet variable: NER SCF IRD REMR SMR All Depedet variable: REMR SCF IRD NER SMR All Depedet variable: SMR SCF IRD NER REMR All

An Empirical Study on the Contribution of Foreign Trade to the Economic Growth of Jiangxi Province, China

An Empirical Study on the Contribution of Foreign Trade to the Economic Growth of Jiangxi Province, China usiess, 21, 2, 183-187 doi:1.4236/ib.21.2222 Published Olie Jue 21 (http://www.scirp.org/joural/ib) 183 A Empirical Study o the Cotributio of Foreig Trade to the Ecoomic Growth of Jiagxi Provice, Chia