Determinants of volatility of net capital inflows to thailand

WILLIAMS COLLEGE LIBRARIES Your unpublished thesis, submitted for a degree at Williams College and administered by the Williams College Libraries, will be made available for research use. You may, through this form, provide instructions regarding copyright, access, dissemination and reproduction of your thesis. _ The faculty advisor to the student writing the thesis wishes to claim joint authorship in this work. In each section, please check the ONE statement that reflects your wishes. 1. PUBLICATION AND QUOTATION: LITERARY PROPERTY RIGHTS A student author automatically owns the copyright to hislher work, whether or not a copyright symbol and date are placed on the piece. The duration of U.S. copyright on a manuscript--and Williams theses are considered manuscripts--is the life of the author plus 70 years. _ I1we not choose to retain literary property rights to the thesis, and I wish to assign them immediately to Williams College. Selecting this option will assign copyright to the College. This in no way precludes a student author Ii'om later publishing his/her work: the student would. however, need to contact the Archives for a permission form. The Archives would be free in this case to also grant permission to another researcher to publish small sections from the thesis. Rarely would there be any reason for the Archives to grant permission to another party to publish the thesis in its entirely: if such a situation arose, the Archives would be in touch with the author to let them know that such a request had been made. _I1we wish to retain literary property rights to the thesis for a period of three years, at which time the literary property rights shall be assigned to Williams College. Selecting this option gives the author a fevv years to make exclusi ve use of the thesis in up-coming projects: articles, later research, etc. / I1we wish to retain literary property rights to the thesis for a period of ~ years, or until my death, whichever is the later, at which time the literary property rights shall be assigned to Williams College. Selecting this option allows the author great nexibility in extending or shortening the time of his/her automatic copyright period. Some students are interested in using their thesis in graduate school work. In this case, it would make sense for them to enter a number such as' lO years' in the blank, and line out the words 'or until my death, whichever is the later.' In any event, it is easier for tbe Archives to administer copyright on a manuscript if tbe period ends with the individual's death--our staff won't have to search for estate executors in tbis case--bm this is entirely up to each student. II. ACCESS The Williams College Libraries are investigating the posting of theses online, as well as their retention in hardcopy. ./ Williams College is granted permission to maintain and provide access to my thesis in hardcopy and via the Web both on and off campus. Selecting this option allows researchers around the world to access the digital version of your work. _ Williams College is granted permission to maintain and provide access to my thesis in hardcopy and via the Web for on-campus use only. Selecting this option allows access to the digital version of your work from the on-campus network only. _ The thesis is to be maintained and made available in hardcopy form only. Selecting this option allows access to your work only from the hardcopy you submit. Such access pertains to the entirety of your work. including any media that it comprises or includes. III. COPYING AND DISSEMINATION Because theses are listed on FRANCIS, the Libraries receive numerous requests every year for copies of works. IfIwhen a hardcopy thesis is duplicated for a researcher, a copy of the release form always accompanies the copy. Any digital version of your thesis will include the release form. / ' Copies of the thesis may be provided to any researcher. Selecting this option allO\vs any researcher to request a copy from the Williams College Libraries. or to make one from an electronic version. _ Copying of the thesis is restricted for _ any researcher. years, at which time copies may be provided to This option allows the author to set a time limit on copying restrictions. During this period. an electronic version of the thesis will be protected against duplication. _ Copying of the thesis or portions thereof, except as needed to maintain an adequate number of research copies available in the Williams College Libraries, is expressly prohibited. The electronic version of the thesis will be protected against duplication. Selecting this option allows no reproductions to be made for researchers. The electronic version of the thesis will be protected against duplication. This option does not dis-allow researchers from reading/viewing the work in either hardcopy or digit~d form. Signed (student author)_ Signatures Removed Signed (faculty advisor) _ v Thesis title 1)e+erMIY\Q11t S of Date tv1 C{ Y I q I - __ VolQhli t-y or: fi-et Cep/ok! ~ fl-z.CtII~ 0'0 'b' Signature Removed _ A ccepted f or theLl rarles __ Date accepted :f"'f/ow) ()_'v :: .: ~--'r--'·CZ="" .:-i ~-,,-,--(Jt_, L.-) %.!;j-' _ _ DETERMINANTS OF VOLATILITY OF NET CAPITAL INFLOWS TO THAILAND by Pakinee Banchuin, Author Professor Peter Pedroni, Advisor A thesis submitted in partial fulfillment of the requirements for the Degree of Bachelor of Arts with Honors in Economics WILLIAMS COLLEGE Williamstown, Massachusetts May 19,2008 Acknowledgements This thesis would not have been possible without the generosity of my advisor, Professor Peter Pedroni, who devoted time and patience to guiding me through the process. I would also like to thank my second reader, Professor Savaser, who took her time to read my draft and give me useful comments. Special thanks go to my friends whom without them, the thesis would have not progressed to this point: Troung Pham Hoai Chung and Aom Kitichaiwat for insisting on reading my drafts, providing constructive comments, and motivating questions; Pablo Cuba and Cid Vipismakul for assistance in computer programming and miscellaneous help; Megan Brankley and Tengjian Khoo for proofreading my drafts. Athikaset T hongves for your helpful overview of Thai economy; and last but not least, Nora Wong for working through late night weekends with me and your wonderful help on the poster. Table of Contents 1. Introduction 2. Literature Review " 2.1 Conceptual Issues 2.2 Suggested Determinants of Capital Inflows 2.3 Empirical Method 3. Model Construction 4. Empirical Studies 4.1 Influence of Push-Pull Factors on Inflows 4.2 Identifying Components of Pull Factors 5. Empirical Results Analysis 6. Conclusion Bibliography Appendix A , Appendix B Appendix C , Appendix D Appendix E , 5 12 28 36 .47 .47 50 54 56 63 66 86 94 1. Introduction Financial integration allows capital to move across countries without any barriers. Capital flows across national boundaries to maximize return on investments and diversify investors' risks. In tum, the free movement of capital increases the availability of funds in emerging economies, enabling these countries to finance their development. One type of capital inflows is foreign direct investment (FDI). When investors build new factories, they usually bring in advanced technology, leading to a transfer of knowledge. On the other hand, capital inflows not always generate positive effects on local economies. More inflows can mean higher demand on the national currency, which induces an appreciation in the exchange rate. This appreciation hurts the country's export competitiveness. Consequently, the inflows incite economic growth in the short run, but since they might hurt export-led growth, the inflows could harm the country's growth in the long run. Capital inflows can also cause inflation. Moreover, the flows place a threat on macroeconomic stability because of a possibility of a sudden reversal of capital flows. A sudden reversal of capital flows is when there is an abrupt stop in capital inflows following by a large increase in capital outflows. Reversals thus drain liquidity in the economy, freezing funds for investment and destroying investor optimism. Ideally, financial integration should maximize the efficient use of resources and the transnational flows equate prices and returns on traded assets in the integrated areas. In reality, there are other factors that affect prices and returns on assets, and the allocation of funds. There are some literature tries to answer these questions but many of these studies use data from the United States and Western European countries which have strong economies, more developed rules of law, and better structured financial markets. These fundamentals are different from those of East Asian countries. The majority of East Asian countries opened their countries to international finance at the same time as developing their capital markets and deregulating their banking sectors. For some of the countries, especially Indonesia, South Korea and Thailand, opening up their capital accounts when their banking systems and financial markets did not measure up to Western standards led to the 1997 economic crisis. Due to the extensive scope of this issue and limited time, this paper only features one developing country. We choose Thailand because it has experienced rapid changes, both an impressive boom and devastating bust, over the last 20 years. The other reason is that some economic literature claims that if it Thailand had not triggered the 1997 economic crisis, the crisis might not have happened at all. One of the causes of the economic crisis was a sudden reversal in capital flows. Nine years later, the Bank of Thailand claimed that Thailand is on the verge of another sudden reversal in capital flows. One signal for the sudden reversal is that after taking into account of all known factors, the Thai Baht still appreciated at a significantly higher rate than those of the regional countries. This implies that the high level of inflows might be because of speculation. To prevent a sudden reversal in capital flows, the Bank launched a 30% capital control to discourage any inflows with the purpose of speculation. Whether those high inflows were hot money for speculation, no one will know for certain, but the actual effect of the capital control was a plunge in Thailand's stock market and an uncertainty in investing in Thailand due to the fear that the Thai government could launch even more extreme policies. The fact that Thailand still faces a risk of a sudden reversal of capital flows indicates that perhaps the driving force behind the flows has not changed even after several post-crisis reforms. Since the data on inflows are not available for all needed time periods, this paper will use the data on net capital inflows (foreign liabilities) instead. This will alter the scope of the study. Instead of investigating the determinants of volatility of the inflows, we will examine the determinants of the fluctuations of the net inflows. Since Thailand has a significantly higher risk of capital flows than developed countries, it is interesting to compare the conclusions on determinants of net capital inflows in Thailand to that data from Western European countries and the States. Thailand experienced high growth rates in the late 1980s and the early 1990s. Asset price bubbles and inadequate bank regulation led to strong capital inflows. Unlike some of its neighboring countries, the majority of its inflows were in the form of bank loans rather than foreign direct investment. Due to the belief that the Thai government would keep the exchange rates fixed, these banks and financial companies did not cover the risk of foreign exchange rates in their loans. Moreover, they used these foreign liabilities to finance long-term projects. Therefore, when there was a sudden reversal of capital inflows in 1997, these firms faced a credit crunch, which finally led to the 1997 East Asian economic crisis. A little over a decade since the crisis, Thailand's economy has recovered to some extent. After the inflows plummeted to the lowest point at negative 15 percent of GDP in the first quarter of 1999, capital inflows slowly bounced back and were at about percent of Thailand's GDP in 2005(Ahuja, Chuenchoksan, and Thaicharoen 2007). The distribution of the flows' components between the pre- and post- crisis periods is different. After the crisis, the main component of capital inflows shifted from bank loans to equity investments. Foreigners' holding of Thai equity reached 65% of total liabilities (Ashvin et al. 2007). Since Thailand and the rest of the world have experienced some rapid and drastic changes over this past decade, it is worth investigating what are the determinants behind the flows which lead to the change in the flows' distribution. Since the volatility of capital flows is associated negatively with macroeconomic stability, knowledge about the determinants of fluctuations in the flows will lead to a better monitoring of macroeconomic stability. To understand the dynamic relationship between capital inflows' volatility and their determinants, this paper will use two Structural Vector Autoregressions (SVAR). The remainder of the paper contains five sections: Section II summarizes the relevant literature. Section III examines the Mundell-Flemming model and places it in the context of Thailand by using quarterly data from Thailand's economy from 1984-2007. Section IV establishes identification matrices to use in VAR. Section V analyzes the empirical results and section VI concludes with key findings. 2. Literature Review 2.1. Conceptual Issues This paper will categorize the determinants of capital inflows into two groups, "push" and "pull" factors. "Push" and "pull" factors are classified by their origins. "Push" factors reduce the attractiveness of lending to developed-country debtors. A common "push" factor is a decrease in rates of return on assets whose fluctuations are usually caused by cyclical factors. Since the rates of return in developed countries fluctuate in response to business cycles, capital inflows motivated by such returns have high volatility. If this kind of capital inflow comprises a high proportion of a country's total inflows, the country is at risk of a sudden reversal of capital inflows (Montiel and Reinhart 1999). Examples of these kinds of "push" factors are the burst of asset bubbles in Japan in the late-1980s and an interest cut in the United States in response to the 1990-91 recession. These two events were concurrent with high level of capital inflows in emerging Asian countries. Two other common "push" factors are growth in an investor base and a need for portfolio diversification. The number of investors has significantly grown, and these investors, particularly of mutual and pension funds, want to diversify their investments, so they put some of their money in developing markets (Taylor and Sarno 1997). Since "push" factors are created outside developing countries, the governments of developing countries cannot curb the factors directly. The government can only implement cautious domestic policies. The other determinants of inflows are "pull" factors, which are country-specific. They improve the risk-return characteristics of assets issued by developing-country debtors (Montiel and Reinhart 1999). For example, attractive PIE ratios in the stock market in emerging countries draw foreign investors. Removing distortions that have spgraph(done) ENDDO i ****check if the restrictions are binding**** com atitle = "Post-crisis: US interest rate and "+ varlabels(i) + '(Reduced form VAR)' spgraph(vfields=4,hfield=1,header=atitle) graph(style=symbols,patterns,header="Impuses Responses of US output" ,key=right,klabels=ilab1,nodates) #Bl(l,l) #Bl(l,2) com atitle = 'Impulse Response Of '+ varlabels(i) graph(style=symbols,patterns,header=atitle,key=right,klabels=ilab1,nodates) #Bl(2,1) #Bl(2,2) spgraph(done) 118 ***6-var- identification matrix*** ***using datafrom 1984:01 -1997:02 cal 1984 14 allocate 1997:02 open dataf:\desktop\08pb\senior thesis\spring\data for rats data(format=rats,org=obs) ***build a vector of series for easier reference*** dec vect[series] var(6) com var = lirusint , tgdp, fdir , tinf , portl, rail diff rusint / drusint diff tgdp / dtgdp diff fdir / dfdir diff tinf / dtinf diff portl / dport diff / dra dec vect[series] dvar(6) com dvar = Iidrusint, dtgdp , dfdir, dtinf, dport, drall ***deseasonalize*** seasonal seasons *Since the amount of GDP, total inflows, and portfolio inflows might vary depending on the time of the year . linreg(noprint) tgdp / destgdp #constant seasons{ -2 to O} linreg(noprint) tinf / destinf #constant seasons {-2 to O} linreg(noprint) portl / desport #constant seasons {-2 to O} *linreg(noprint) / desport *#constant seasons{-2 to O} ***difference the deseasonalized diff destgdp / ddestgdp diff destinf / ddestinf diff desport / ddesport ***checking for unit root*** *portl commlag=12 llags=mlag, 1,-1 119 linreg ddesport / #constant desport{1} ddesport{1 to Hags} compute mtratio = %beta(Hags+1)/sqrt(%seesq*%xx(Hags+1,Hags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if Hags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg ddesport # constant desport{ 1} dport{ to maxlag} *t-stat = -1.91, sig. level = 0.06 *tgdp com mlag=12 Hags=mlag,1,-1 linreg ddestgdp / #constant destgdp{ 1} ddestgdp{ to Hags} compute mtratio = %beta(llags+1)/sqrt(%seesq*%xx(Hags+1,Hags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=Hags ; break;} end Hags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg ddestgdp # constant destgdp{ 1} ddestgdp{ to maxlag} *t-stat = -0.42; sig level = 0.68 *fdir com mlag=12 llags=mlag,1,-1 linreg dfdir / #constant fdir{ 1} dfdir{ to llags} compute mtratio = %beta(llags+1)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg dfdir # constant fdir{l} dfdir{ to maxlag} 120 *t-stat = -3.69, sig-lvel = 0.00 *tinf com mlag=12 llags=mlag,1 ,-1 linreg ddestinf / #constant destinf{ I} ddestinf{ to llags} compute mtratio = %beta(llags+1)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= 'maxlag linreg ddestinf # constant tinf{ I} ddestinf{ to maxlag} *t-stat =0.24, sig. level = 0.81 *ra com mlag=12 llags=mlag,l,-l linreg dra / #constant ra{I} dra{I to llags} compute mtratio = %beta(llags+l)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= 'maxlag linreg dra # constant ra{I} dra{I to maxlag} *t-stat = -1.58, sig. level = 0.12 ***find appropriate lag lengths for the VAR system COMPUTE mlag = 12 DO lag= 1,mlag SYSTEM to VARS drusint ddestgdp dfdir ddestinf ddesport dra lags to lag det constant end(SYSTEM) ESTIMATE(noftests) COMPUTE SBIC= %nobs*%logdet+%nregsystem*log(%nobs) 121 DISPLAY 'lag' lag 'SBIC=' sbic END DO lag ***sbic informs us that 10 lags is the most appropriate, *but at 10 lags, there are no degree of freedom. So we use lags. ***estimate VAR and change to SVAR system(model=modell) to vars drusint ddestgdp dfdir ddestinf ddesport dra lags to det constant end(system) estimate(coeff=c 1,outsigma=v1) / COMPUTE Rl = %VARLAGSUMS COMPUTE OMZ = INV(Rl)*vl *tr(inv(Rl)) COMPUTE Al = %DECOMP(OMZ) COMPUTE AO = Rl *Al ***Plot Impulse Responses and Variance Decompositions*** ***check if restrictions are binding*** declare vect[string] slabels(6) com slabels = II'Real Push factors shock' , 'Thai supply shock' , 'Thai risk premium shock' , 'Thai general real demand shock' , 'Thai nominal shock' , 'Nominal push factor shock'lI declare vect[string] varlab(6) com varlab = II'Foreign real interest rate' , 'Thai GDP' , 'FDI as share of capital flows' , 'Total capital inflows' , 'Portfolio inflows' , 'Reserve assets'li j=I,6 spgraph(vfield=6, hfield=l, header='Pre-crisis: Impulse Responses to '+ slabels(j) + ' (differences)') i=I,6 graph(nodates, header='Impulse Response of' + varlab(i)) #Bl(i,j) end i spgraph(done) end j declare rect[series] SBl(6,6) declare rect[series] SEl(6,6) declare rect[series] SEA(6,6) i=I,6 j=I,6 acc Bl(i,j) steps SBl(i,j) ace El(i,j) steps SEl(i,j) set SEA(i,j) = SEl(i,j)/number end j; end i= 1.64 {; compute maxlag=llags ; break;} end llags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg ddesport # constant desport{I} dport {I to maxlag} *t-stat = -1.79 *tgdp commlag=12 llags=mlag,l,-l linreg ddestgdp / #constant destgdp{ I} ddestgdp{ to llags} compute mtratio = %beta(llags+ l)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if llags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg ddestgdp # constant destgdp {I} ddestgdp {I to maxlag} *t-stat = -1.75 *fdir com mlag=12 131 llags=mlag,l,-l linreg dfdir / #constant fdir{ I} dfdir{ to llags} compute mtratio = %beta(llags+ l)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtrati0) >= 1.64 {; compute maxlag=llags ; break;} end Hags if Hags == .and. abs(mtratio) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= 'maxlag linreg dfdir # constant fdir{ I} dfdir{ to maxlag} *t-stat = -0.51 *tinf com rnlag=12 llags=mlag,1 ,-1 linreg ddestinf / #constant destinf{ I} ddestinf{ to Hags} compute mtratio = %beta(llags+ l)/sqrt(%seesq*%xx(llags+1,llags+1)) if abs(mtratio) >= 1.64 {; compute maxlag=llags ; break;} end llags if Hags == .and. abs(mtrati0) < 1.64 {; compute maxlag = ;} end if disply 'maxlag= ' maxlag linreg ddestinf # constant tinf{ I} ddestinf{ to maxlag} *t-stat =-2.33 ***find appropriate lag lenths for the VAR system COMPUTE mlag = 12 DO lag= 1,mlag SYSTEM to VARS drusint ddestgdp dfdir ddestinf ddesport lags to lag det constant dum2 end(SYSTEM) ESTIMATE(noprint, noftests) COMPUTE SBIC= %nobs*%logdet+%nregsystem*log(%nobs) DISPLAY 'lag' lag 'SBIC=' sbic END DO lag ***sbic informs us that lags is the most appropriate, but at lags, there are no degree of freedom. So we use lags which gives us 28 degrees of freedom. ***estimate VAR and change to SVAR 132 system(model=modell) to vars drusint ddestgdp dfdir ddestinf ddesport lags to det constant end(system) estimate(coeff=c 1,outsigma=v1) I COMPUTE Rl = %VARLAGSlJMS COMPUTE OMZ = INV(Rl)*vl *tr(inv(Rl» COMPUTE Al = %DECOMP(OMZ) COMPUTE AO = Rl *Al ******************Plot Impulse Responses and Variance Decompositions****************** compute steps=16 impulses(model=modell ,results=B 1,decompose=AO,noprint) * steps * errors(noprint,model=modell ,results=El ,decompose=AO) * steps * i = 1,2; j=I,2 pri / B 1(i,j) pri / El(i,j) end j=I,2; end i =1,2 declare rect[series] SBl(5,5) declare rect[series] SEl(5,5) declare rect[series] SEA(5,5) i=I,5 j=I,5 acc Bl(i,j) steps SBl(i,j) acc El(i,j) steps SEl(i,j) set SEA(i,j) = SEl(i,j)/number end j; end i[...]... Considering the pattern of net total inflows, its volatility started to increase since 1990 26 After the crisis, the level of net total inflows plummeted to the lower level than where its pre-crisis value was This is similar to the net inflows of other investments and somewhat close to that of portfolio investment This suggests that the movement of net total inflows in Thailand is largely motivated by... If the major determinants of capital inflows are domestic, the government might be able to control the sustainability of capital inflows If push factors are the major determinants of fluctuations in capital inflows, the government can implement sterilization or fiscal policies to cope with the effect of the 6 flows Sterilization policies for capital adjust the domestic component of the monetary base... of total inflows and specific types of inflows In Part II, the analysis will be about identifying the influence of each component in pull factors on different types of inflows 4.1 Influence of Push-Pull Factors on Inflows We will use short run restrictions about push-pull factors to obtain an identification matrix A Push factors Since capital usually flows from countries with big open econOIllies to. .. Thai interest rate Since Thailand is a small open economy, an increase in Thailand' s interest rate does not induce a rise in the interest rate of the "rest of the world" Higher interest rate in Thailand will attract more investors to move their capital to Thailand The corresponding IS-LM equations and diagrams are in appendix A 25 The proposed determinants of volatility of net inflows from economic literature... the level of capital inflows 24 I An increase in the interest rate ofthe "rest ofthe world" An increase in the interest rate of the "rest of the world" induces a nse In Thailand' s interest rate Since there is a risk premium in Thailand, the size of increases in interest rates in both places need not be equal Depending on their risk preferences, investors might move their capital into or out of Thailand, ... enable the government to have certain level of control over the size of inflows 7 2.2 Suggested Determinants of Capital Inflows In the last fifteen years, growing capital flows from developed countries to developing countries have contributed both to economic growth and to economic crises The main focus in the literature on this phenomenon is to identify whether "push" factors (external to the representative... change in capital flows We rely on Thai historical data on net inflows for the relative importance and characteristics of each type of net inflows to Thailand (Graphs are shown in Appendix B) Thailand slowly opened up its capital account in the mid-1980s In 1993, Thailand established the Bangkok International Banking Facilities (BIBF) The BIBF led to two important consequences: the liberalizing of Thai... factors (domestic) are the major determinants of capital inflows Some findings show that push factors have greater effects on capital inflows Calvo et al (1993) find that an increase in interest rates abroad leads to a rise in capital outflows from Latin America Furthermore, push factors significantly influence the movements of official reserves and real exchange rates, yet the impact of push factors... Since the characteristics of other investment are difficult to identify and the mobility of portfolio inflows can easily lead to a sudden reversal of capital inflows, in this study, we will pay special attention to the net inflows of portfolio investment 27 4 Empirical Studies Since capital inflows and their associated variables are dynamic, we use a structural VAR analysis to test the predictions Impulse... the institutional investor credit ration affect the inflows into the group of Asian countries in their paper Guided by these findings, we will categorize 8 detenninants of capital inflows to Thailand into push and pull factors After that, we will try to isolate subcategories of push and pull factors 2.3 Empirical Method We use Structural Vector Autoregression (SVAR) developed by Blanchard and Quah . major determinants of capital inflows are domestic, the government might be able to control the sustainability of capital inflows. If push factors are the major determinants of fluctuations in capital. data on net capital inflows (foreign liabilities) instead. This will alter the scope of the study. Instead of investigating the determinants of volatility of the inflows, we will examine the determinants. (Jt_, %.!;j-' _ DETERMINANTS OF VOLATILITY OF NET CAPITAL INFLOWS TO THAILAND by Pakinee Banchuin, Author Professor Peter Pedroni, Advisor A thesis submitted in partial fulfillment of the requirements