FOREIGN INSTITUTIONAL INVESTOR S IMPACT ON STOCK PRICES IN INDIA

Transcription

1 FOREIGN INSTITUTIONAL INVESTOR S IMPACT ON STOCK PRICES IN INDIA ANAND BANSAL Punjabi Universiy Guru Kashi Campus Damdama Sahib-530, Punjab Phone: ; Fax: preemillie@yahoo.com J.S. PASRICHA Deparmen of Commerce Punjabi Universiy Paiala-4700-Punjab Phone: ; Fax: jspasricha@rediffmail.com Absrac This paper sudies he impac of marke opening o FIIs, on Indian sock marke behavior. India announced is policy regarding he opening of sock marke o FIIs for invesmen in equiy and relaed insrumens on 4 h Sepember 99. Using sock marke daa relaed o Bombay Sock Exchange, for boh before and afer he FIIs policy announcemen day. An empirical examinaion has been conduced o assess he impac of he marke opening on he reurns and volailiy of sock reurn. We found ha while here is no significan changes in he Indian sock marke average reurns, volailiy is significanly reduced afer India unlocked is sock marke o foreign invesors. Keywords: Foreign Insiuional Invesors; Sock Prices Reurn; Sock Prices Volailiy; Foreign Invesmens.. INTRODUCTION The remarkable economic growh during he pas wo decades in mos of he emerging counries had been simulaed by foreign capial inflows from developed counries. The pos 990s period winessed sharp augmen in flows of foreign privae capial and official developmen finance los is predominance in ne capial inflows. Mos of he developing counries opened heir capial markes o foreign invesors eiher because of inflaionary pressures, widening curren accoun deficis, exchange depreciaion, increase in foreign deb or as a resul of economic policy. There was a surge in capial inflows ino India oo since 99 as VOLUME NUMBER OCTOBER 009 8

2 in India, he purchase of domesic securiies by FIIs was firs allowed in Sepember 99 as par of he liberalizaion process ha followed he balance of paymen crisis in (Gordon and Gupa, 003). Now days, a significan porion of Indian corporae secor s securiies are held by Foreign Insiuional Invesors, such as pension funds, muual funds and insurance companies. These invesors are ofen viewed as sophisicaed invesors as hese insiuional invesors are beer informed and beer equipped o process informaion han individual invesors (Han and Wang, 004). As he share of foreign invesors in emerging markes has risen, hey have influenced he asses prices considerably. Consequenly, policymakers have become increasingly concerned abou he facors deermining inernaional invesmen, he performance of foreign capial invesmens, and he impac of foreign invesmen on local urnover and on he volailiy of sock prices (Tesar and Werner, 995). The impac sudy of FIIs flows on domesic sock marke is imporan from governmen as well as invesor poin of view, for example, does he opening up of he marke for FII increase speculaion in he marke and hus make he marke more volaile and more vulnerable o foreign shocks (Li, 00). The immediae impac of marke opening o FIIs is he surge in rading volume and capial inflows o domesic sock markes, resul of which he boom in sock prices. The sock marke boom, ypically, does no las for he enire period is of capial inflows. I usually sars wih he iniial surge in capial inflows and ends before he episode of capial inflows compleely subsides (Calvo and Mendoza, 000). Henry (000) repors he wo possible consequences of marke liberalizaion in he ligh of inernaional asse pricing models. Firs oucome of marke liberalizaion (because of is impac on he cos of capial) is an increase in a counry s equiy prices because marke learns ha domesic markes will liberalize more in near fuure. The second impac of marke liberalizaion is on physical invesmen ha will increase because of fall in cos of capial as new enrepreneurs will iniiae more invesmen projecs. The second effec of marke liberalizaion will definiely increase he rae of economic growh. Similarly Gompers e al. (00) prove ha insiuional invesors invesed in liquid and large socks having low reurns during he previous year. So an increase in he insiuional demand in share marke will affec sock marke prices and reurns if supply and demand curves for ha paricular share are no perfecly elasic. Han e al. (004) also analyze he impac of insiuional invesors on sock prices from a differen perspecive. They sudied he impac on sock prices because of he invesmen consrains on insiuional invesors by heir uni holders. Insiuional consrains some ime refrain from selling or purchasing of sock abou which hey have even some good/bad news. So hey conclude ha higher insiuional invesmen consrains have srong price momenum on he shares. Similarly Lin e al. (006) conclude ha he invesmen performance of FIIs high holding socks is significanly beer han ha of FIIs low holding socks. They presened he evidence ha FIIs rading behavior has generaed beer reurns and porfolio performance since he sock marke s full liberalizaion. Li (00) sudies he impac of marke opening o foreign invesors on Taiwan sock marke behavior and found no significan changes in sock marke reurn afer marke opening. Bu auhor agreed ha he impac on reurn should be here because large inernaional 8 VOLUME NUMBER OCTOBER 009

3 invesors end o sudy companies more horoughly. The involvemen of foreign invesors disseminaes informaion beer hence leads o more efficien marke. Richards (004) analyze daa of six Asian emerging equiy markes and found wo ineresing findings. The rading behavior of foreign invesors was largely influenced by he reurn in global marke ha is posiive feedback rading. The price impac associaed wih foreign invesors rading was much large han esimaed earlier. In he presen paper we look a he impac of FIIs enry on he sock marke behavior. For his purpose, changes in marke reurn and volailiy have been examined. Secion of his paper will inroduce he daa and mehodology. Secion 3 presens he oucomes of sudy and is followed by cerain final observaions and remarks in Secion 4.. DATA AND METHODOLOGY Bombay Sock Exchange (BSE) is he only surviving oldes exchange in India. BSE is considered as he baromeer of Indian economy. The daa relaed o is prominen marke index, SENSEX (consising of 30 blue chip socks) has been used in his paper o empirically compare he marke behaviour before and afer he Indian marke opening day (he even day). Exclusively we sudy he change of marke reurn and volailiy afer he enry of FIIs o Indian capial marke. Though FIIs were allowed o rade in he Indian sock marke from 4 h Sepember 99 bu hey made firs invesmen in he monh of January 993. Thus he 4h Sepember 99 has been chosen as even day because here would have been a boos in marke aciviy upon he announcemen regarding marke opening o FIIs. We used he marke index daa from 3 rd January 99 o 9 h March 994. Our daa se covers abou wo years period ha included daa of 330 rading days before he even day and 330 rading days afer he even day. Period before even dae Period afer even dae Name of Sock Exchange Index used o calculae daily reurn and volailiy Frequency of Daa 3 rd January 99 o h Sepember 99 (Toal 330 rading days) 5 h Sepember o 9 h March 994 (Toal 330 rading days) Bombay Sock Exchange BSE SENSEX Daily Table : Descripion of daily daa used in sudy VOLUME NUMBER OCTOBER

4 .. STOCK PRICES RETURN To evaluae he impac of Indian sock marke opening o FIIs on sock prices, average reurn before and afer he even day has been calculaed for differen sub sample days. The reurn has been calculaed by aking he difference in he naural logarihm of he closing index values for wo consecuive rading periods. Symbolically, he rae of reurn has been calculaed as follows: P R = log e 00 () P R sands for he rae of reurn for a given period, where log e () is he naural logarihm, P is he closing value of sock index (SENSEX in our case) on dae and P is he closing value of sock index. On he basis of daily reurn, average marke reurn has been calculaed for k rading days before and k days afer he even day. If he average reurns for k days before and k days afer even days are represened by ( r, r,..., r k ) and ( r, r,..., r k ) respecively, we can es he null hypohesis of equal average reurn as: Null Hypohesis H 0 : µ = µ () (There is no difference in Indian sock prices reurn before µ and afer µ he enry of FIIs).. STOCK PRICE VOLATILITY The change of volailiy in he Indian sock prices has been examined by comparing he variance of he reurns of sub sample days before and afer he even day. Sandard deviaion is he saisical ool used o measure he volailiy in he reurns of various markes. Thus, volailiy can be calculaed as: S.D = ( r r) (3) n Here, n is he number of observaions (no of rading days in a sub sample), r is he daily rae of reurn in a sub sample, r is he average reurn in a sub sample. Le σ i denoe he variance of boh periods ha is before and afer even, hen he hypoheses o be esed can be wrien as: Null Hypohesis H 0 : σ = σ (4) (There is no difference in sock prices volailiy before σ and afer σ he enry of FIIs). 84 VOLUME NUMBER OCTOBER 009

5 3. EMPIRICAL RESULTS 3.. IMPACT OF FIIS INVESTMENT FLOWS ON STOCK PRICES RETURN: The calculaed average reurns for before and afer sub samples are repored in able II. Wilcoxon-Mann-Whiney es, also called he rank sum es or U es, is he measure which has been used o check he significance of differences in he average marke reurns of he all sub sample wih same and + even days. Wilcoxon-Mann-Whiney es is used o es he null hypohesis ha wo sub samples are idenically disribued or no. The Mann Whiney U saisic is defined as: U n n + n n + ( ) R (5) = i (Here n and n are sample size and Ri is sum of ranks) The mean of U saisic is: µ = n n ) / (6) U ( n n ( n + n + ) Sandard error (or sandard deviaion): σ = (7) If he sample size is more han 8, wih he given level of saisical significance, he es will repor wheher he calculaed value of U es saisic falls wihin he accepance region or no. The U es saisic can furher be inerpreed by using he Z saisic as Z is a sandard normal deviae whose significance can be checked from ables of he normal disribuion. If he calculaed Z (ignoring signs) value does no equal or exceed he criical Z value of.96 (criical Z value for a wo-ailed es a 5%), hen i can be assumed ha he null hypohesis is correc and here is no difference beween wo groups of sample. However if he Z value exceeds.96 hen here is sufficien evidence o rejec he null hypohesis and accep he alernaive hypohesis. Reurn Significance Reurn Afer Z Before Level ± ± ± ± ± ± ± ± ± ± ± Table : Tesing he Average Reurn Change VOLUME NUMBER OCTOBER

6 The daily mean reurn for he 30 days, prior o he enry of FIIs has been calculaed a 0.866, while for he 30 days period following he even day is When he mean reurn for 60 days before and afer FIIs enry o India are mached, he mean reurn following FIIs enry day are found declining from o Similarly for he pos FIIs enry sub samples of 90, 0, 50, 80, 0, 40 days have shown negaive reurns as compared o posiive reurns for hese sub sample days before he enry of FIIs. From 70 days onward afer enry, reurn has again become posiive bu sill found significanly less han he reurn before he FIIs enry for he same periods. Figure : Sock prices reurn before and afer he enry of FIIs Using he Mann-Whiney Tes (Wilcoxon Rank Sum Tes) and a significance level of α = 0.05 (5%), here is enough evidence o conclude ha here is no significan difference in mean daily reurn before and afer he announcemen dae for he enry of FIIs in Indian marke. Evidence is srongly in he favor of null hypohesis as he wo ail probabiliy levels (0.055, 0.34, and so on) for all he sub sample is more han 0.05 and he values of Z (ignoring signs) are also less han.96 (abulaed values of Z a 5%). We have o accep our null hypohesis ha is here is no difference in Indian sock prices reurn before ( µ ) and afer µ ) he enry of FIIs. ( 3.. IMPACT OF FIIS INVESTMENT FLOWS ON STOCK PRICES VOLATILITY: The change in sock price volailiy level has been sudied by comparing he variance of reurns, before and afer he enry of FIIs. Our null hypohesis for esing is σ = σ and alernaive hypohesis is σ σ, where σ is he variance before he announcemen, and σ is he variance afer he announcemen of FIIs policy in India. Levene (960) proposed he saisical es of equaliy of variance which has been used here for sudy. The Levene s es uses he average of he absolue deviaions, insead of he mean square of deviaions, so he avoidance of 86 VOLUME NUMBER OCTOBER 009

7 squaring makes he es crierion much less sensiive o non-normal disribuions (Chukwuogor and Chiaku, 007). Lieraure suggess ha no assumpion of normaliy is required while using Levene s es. A low value of significance associaed wih he Levene s es indicaes ha boh sub samples have unequal variances and he null hypohesis is false. The able III presens he volailiy of sock prices reurn for various sub sample periods before and afer he even day. The daily volailiy for he firs sub sample i.e. 30 days prior o he even day has been , while for he 30 days afer he enry of FIIs has been Similarly he subsequen values for 60 days and 90 days are.09, and.7555, respecively. As i can be seen from Table III, all sub sample sandard deviaions afer he enry of FIIs have been smaller han he corresponding counerpars before he enry of FIIs excep for he 30 days case. Days SD Before SD Afer Levene Saisic Significance Level ± ± ± ± ± ± ± ± ± ± ± Table 3: Tesing he marke volailiy change By using he Levene es, we find ha he volailiy changed significanly for differen periods. The Levene s saisic and is significance levels which have been presened in he able III showed ha firs wo samples were no wihin accepable region a 5% level of significance, bu aferward, he values urned wihin accepable region a 5% significance level. This oucome suggess ha by he enry of FIIs in Indian marke, he marke volailiy did no change for he immediae wo monhs period. However afer wo monhs volailiy reduced significanly, so he null hypohesis of σ = σ sands rejeced in he favor of σ σ. VOLUME NUMBER OCTOBER

8 Figure : Sock prices volailiy before and afer he enry of FIIs The figure here plos he mean volailiy for all he sub sample periods. I can be clearly winessed here ha volailiy afer he passage of 30 days sared declining in comparison o he same period before he enry of FIIs. These resuls are idenical wih he earlier empirical work conduced on he oher emerging markes. For example, Li (00) found ha while here was no significan changes in he sock mean reurns, volailiy was significanly reduced hree monhs afer Taiwan opened is sock marke. Bekaer e al. (997) reveal ha mos counries ha have experienced liberalizaion had reducion in volailiy. However, De Sanis e al. (997) did no find any evidence abou he impac of marke opening on sock reurn volailiy. Kim e al. (000), on he oher hand, found ha sock reurns increased immediaely afer marke opening bu fell subsequenly and here was no accompanying increase in he volailiy of sock reurns. 4. CONCLUSION On he basis of above discussion, i can be said ha while reurn declined reasonably afer he enry of FIIs, he volailiy has been reduced significanly afer heir enry. Besides, FIIs invesmen flows, here may be oher reasons as well ha may have some degree of influence on marke volailiy and reurn. While he FIIs invesmen flows and conemporaneous SENSEX, NIFTY, marke capializaion and marke urnover have been srongly correlaed in India, he correlaion beween FIIs invesmens and marke volailiy and marke reurn has been comparaively low. I means volailiy in Indian marke is no he funcion of FIIs invesmen flows. There may be some oher reasons which induced he volailiy in Indian marke over he ime. 88 VOLUME NUMBER OCTOBER 009

9 5. REFERENCES Bekaer, G. and Harvey, C.R. (997). Emerging equiy marke volailiy. Journal of Financial Economics, Vol. 43, Chukwuogor, Chiaku (007). An economeric analysis of African Sock Marke: Annual reurns analysis, day-of-he-week effec and volailiy of reurns African Journal of Accouning, Economics, Finance and Banking Research, Vol., No., Calvo, G.A. and Mendoza, E.G. (000). Raional conagion and he globalizaion of securiies markes. Journal of Inernaional Economics, Elsevier, Vol. 5, No., De Sanis, G. and Imrohoroglu, S. (997). Sock reurns and volailiy in emerging financial markes. Journal of Inernaional Money and Finance, Elsevier, Vol. 6, No.4, Gompers, P.A. and Merick, A. (00). Insiuional invesors and equiy prices, Quarerly Journal of Economics, Vol. 6, No., 9-59.Gordon, J.P. and Gupa, P. (003). Porfolio flows ino India: Do domesic fundamenals maer, IMF 003, Working Paper No, 03/0. Han, B. and Wang, Q. (004). Insiuional invesmen consrains and sock prices. Dice Cener for Research in Financial Economics 004, Working Paper No, Henry, P.B. (000). Sock marke liberalizaion, economic reform, and emerging marke equiy prices. Journal of Finance, American Finance Associaion, Vol. 55, No., Kim, E. H. and Singal, V. (000). Sock marke openings: Experience of emerging economies. Journal of Business, Universiy of Chicago Press, Vol. 73, No., Levene, H. (960). Robus ess for equaliy of variances. In I.Olkin (Ed.), Conribuions o probabiliy and saisics (pp. 78-9), Palo Alo, California: Sanford Universiy Press. Li, Q. (00). Marke opening and sock marke behaviour: Taiwan s experience. Inernaional Journal of Business and Economics, Vol., No., 9-5. Lin, A. and Chen, C.Y. (006). The impac of qualified FII on Taiwan's Sock Marke. Web Journal of Chinese Managemen Review, Vol.9, No., -7. Richards, A. (004). Big fish in small ponds: The momenum invesing and price impac of foreign invesors in asian emerging equiy markes. Journal of Financial and Quaniaive Analysis, Vol. 40, No.. Tesar, L.L. and Werner, I.M. (995). U.S. equiy invesmen in emerging sock markes. World Bank Economic Review, Oxford Universiy Press, Vol. 9, No., VOLUME NUMBER OCTOBER