Investigation of Relationship between Stock Prices, Interest Rate and Exchange Rate Fluctuations

Transcription

1 Vol. 2 No. 4, 2014, Investigation of Relationship between Stock Prices, Interest Rate and Exchange Rate Fluctuations Amir Haji Ahmadi 1, Tahmineh Sanei Emamgholi 2 Abstract One of the most important matters that has always been considered in the stock market is analysis of the effects of economic condition on stock price changes, and these changes are the prime cause for fluctuations of stock price index. As long as the value of stock in the stock exchange is influenced by various factors particularly macroeconomic variables, this study aims at examining the relationship between stock price on one side and bank interest rate as well as currency rate fluctuations on the other side. In this study, the relationship between stock price and bank interest rate as well as currency rate fluctuations is tested using data collected from 50 distinguished (more active) firms as the sample. The data was analyzed utilizing autoregression model, Johansen test, Augmented Dickey-Fuller test, forecasting error analysis of variance, effect statistic tests, and maximum likelihood estimation. In this research, the time period under study is the years between 2008 and The results of the study revealed that although the real rate of bank interest and market currency rate had negative effects on the index of 50 distinguished firms of the stock exchange, the nominal rate of bank interest and reference currency rate had positive and significant effects on the index of 50 firms of the stock exchange. In other words, testing hypotheses of the study showed that there was a significant correlation between market currency rate and price index of 50 distinguished firms of the stock exchange and that there was also a significant correlation between the real rate of bank interest and price index of 50 distinguished firms of the stock exchange. Keywords: Stock price, Market currency rate, Reference currency rate, Real rate of bank interest, Bank interest rate. 1. Introduction Financial market is divided into two parts namely capital market and money market. Although these two markets have clear definitions and are independent from each other, they have interactions at the same time. Therefore, policy-making in each part may have effect on the other part. In this study, the fluctuations of bank interest and currency rate in money market and their effect on capital market are concerned. In fact the main question is Is there any significant relationship between bank interest rate as well as fluctuations of currency rate and stock price in the concerned firms with the time period under study? Obviously, the capital market which is mainly represented by stock exchange in Iran is influenced by many variables such as privatization, economic and political tensions, fluctuations of currency rate, inflation, stock price, etc. In the present study, we intend to examine the long-term relationship between stock price index and currency rate as well as bank interest. In fact, this study aims at testing the long-term effect of each of the above-said variables on price index of stock in stock exchange in terms of degree and direction. Examining such a relationship may help policy makers with orientations of monetary and currency policies, and determine the effect of these policies on the stock exchange. In this study, the researcher, taking into account the theoretical literature of the subject, intends to examine the effect of currency rate changes as an effective variable, and 1 Member of the partment Management Imam khomeini higher Education center (I h e c), In Iran Albor, Karaj, Iran 2 Master student in non-governmental and non-profit institution Samangan, Amol, Iran 2014 Research Academy of Social Sciences 182

2 also fluctuations and changes of bank interest on stock price in Iran in order to determine to what extent these fluctuations have effect on stock price changes. 2. Review of Literature Ja Weng Hung (2009) examined the causal relationship between the stock price and currency rate in Korean economy during the years In this research, the data was collected from Korean stock market and international financial statistics. To analyze the stability of the two variables, ADF and KPSS tests were used. Subsequently, the causal relationship of the two variables was tested using Engle-Granger Regression analysis. The results revealed that devaluation has negative effect on stock price in the short run. That means changes in currency rate affect exports of firms, and finally they have effect on stock price. Based on these findings, there is a bilateral relationship between stock price and currency rate in the long run, however, in the short run, the same relationship is unilateral. Christopher Gan et al. (2006) tested the interaction between New Zealand stock index and a seven-part set of macroeconomic variables for the monthly data from January 1990 till January 2003 using cointegrate tests. The economic variables included inflation rate, currency rate, GDP, money supply, long-term interest rate, short-term interest rate and domestic retail oil price. The results of Johansen cointegrate test showed that there was a long-term relationship between New Zealand stock price rate and the concerned economic variables. The results of Granger causality test showed that New Zealand stock price rate is not the granger causality for changes in economic variables. Its cause is the small size of New Zealand stock market compared to stock markets of the developed countries. Hessing (2011) researched the relationship between macroeconomic variables and Hungry stock price index for the time period of To this end, he used seasonal data and GCRAH model. The results revealed that the stock price index had positive correlation with GDP and ratio of government s debt to GDP, and that it had negative correlation with effective nominal currency rate, real interest rate and expectable inflation rate. 3. Methodology This study aims at exploring the effects of currency rate ad bank interest on price index of 50 more active firms of the stock exchange. To this end, different methods such as structural equations and equations system were simultaneously used. In this study, the effects of currency rate and bank interest on stock price of 50 firms more active in the stock exchange was examined using VAR model and Johansen long-term relationship that has advantages over structural equations. This method in effect is a linear relationship between the dependent variable and inhibitions of all present variables in equations system. To test the unit root in variables of the model, augmented Dickey-Fuller unit root test was run. To determine the number of optimal inhibitions in this model, Schwarz's Bayesian criterion (SBC) was utilized. ADF test was run with regard to intercept and without process, and Schwarz criterion was used to determine the durability of the variables of the model. For determining the length of optimal inhibitions in the model, academic criteria such as AIC, SBC as well as LR and LR were utilized. To prove and confirm existence of a long-term relationship for the estimated model of endogenous relationship between currency rate as well as real rate of bank interest and price index of 50 distinguished firms in the stock exchange, Johansen method, Trace and Max were used. In this study, where the relationship between currency rate and bank interest rate variables are examined, to explore the statics of time series, unit root test and to estimate the model, VAR was used 3. The statistical population of this study is the firms admitted to Tehran Stock Exchange, and the sample is 50 distinguished firms (more active) in the stock exchange within Sarmad, Z.; Barzgan, A., Hejazi, Mehdi. (2011), Research methods in the behavioral sciences, Informed Publishing, pg

3 Hypothesis A. H. Ahmadi & T. S. Emamgholi Hypothesis I: There is a significant relationship between stock price index in stock exchange and bank interest rate. Hypothesis II: There is a significant relationship between currency rate fluctuations and stock price index in the stock exchange. Variables There are a dependent variable and four independent variables in this study, which are explained below. A) Independent Variables The independent variables in this study are real rate of bank interest (interest rate), nominal interest rate, bank interest rate, dollar rate of free market and reference currency rate of central bank. B) Dependent Variables: Price Index of 50 Distinguished Firms in Stock Exchange The information as to monthly time series concerning data of five-year interest rate ( ) of bank long-term deposit investment, monthly data of inflation rate, monthly data of dollar currency market rate and monthly data of price index of 50 distinguished firms in the stock exchange were collected using information offered by the central bank, census bureau, and website of Tehran Stock Exchange at The data were computed and their results were used as input for Eviews software. 4. Results First, augmented Dickey-Fuller unit root test was run for all variables based on fixed and process component significance. To determine the number of optimal inhibitions in these models, SBC was utilized. The results showed that all variables were unstable at the data level. However, repetition of this test as to data difference indicated that all variables were unstable after one round of difference calculation, and their instability hypothesis is rejected at 5% level. Therefore, all variables of the model are accumulated from 1 st degree or 1(1). The results of the test are displayed in the table below. ADF test was run with regard to intercept and without process, and Schwarz criterion was used to determine the durability of the variables of the model. The results of the test are as follows: Table 1: Results of Augmented Dicky-Fuller Unit Root Test at Data Level & Difference Test statistics at Variable data level Critical values Number of Degree of Test statistics at at 5% level inhibitions accumulation data difference Currency rate (2) Real rate of bank interest (2) Price index of distinguish 1(1) firms Another matter is that if the variable in a VAR model need to be durable? In this regard, Sims (1980) suggested not using difference calculation of variables even if the variables have unit root. He believed that the purpose of VAR analysis is determining the relationship between the variables not estimating parameters. Generally speaking, it is believed that difference calculation of data 4 The Tehran Stock Exchange (Various years) Monthly Statistical 184

4 ruins convergence of data. Moreover, the data need not to be deprocessed. Fuller (1976) showed that if asymptotic efficiency of VAR is concerned, even if difference calculation of data is suitable, no noticeable result is obtained. Furthermore, he believes that not only difference calculation of variables ruins the model information but also no useful result is obtained; therefore, for estimating the model, the concerned variables are used without difference calculation. Determining Optimal Inhibition Length in the Model In principle, there is nothing to avoid inserting a great number of variables in VAR model. In fact, n equation with p inhibitions is feasible for each equation in the system; however, VAR model based on Sims assumption, requires knowing two items, first, variables suitable for the system and second, determining the number of inhibitions. If a different inhibition length is determined for each variable in VAR equation, the system shall not be symmetrical any longer, and OLS estimates will not have the necessary efficiency. It is customary to use similar inhibition length for equations of the system. Sometimes, limitation of degree of freedom determines the number of inhibitions. However, when there are a great number of observations, a suitable criterion should give its place to a better model. Before estimating VAR model, optimal inhibitions of model s variables must be identified. As shown in the table 4-1, to determine optimal inhibition length of the model, AIC, SBC as well as adjusted LR and LR were utilized. AIC and SBC have shown degree 1 to be the optimal inhibition length of the model (maximum value of each of these statistics determines the optimal degree of the model). Based on adjusted LR and LR criteria, the optimal degree of the model is 0. However, in practice, using these criteria does not lead to similar results for determining the optimal degree of VAR model. Since using SBC compared to other criteria loses less degree of freedom, value of 1 is determined for inhibition length of the model based on this criterion. Table 2: Determining Number of Optimal Inhibitions of Estimated Model VAR Lag Order Selection Criteria Endogenous Variables: firms nexchange nrate rate exchange Exogenous Variables: c Date: 09/02/13 Time: 09:32 Sample: 1387 (2008) M (2012)M12 Included Observations: 49 HQ SC AIC FPE LR LogL Lag e+21 NA * e e e * * 1.28e+17* * * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwartz information criterion HQ: Hannan-Quinn information criterion The estimated model in this article shall be as follows: FIRMS t =α+β 1 EXCHANGE t +β 2 RATE t +β 3 NEXCHANGE t +β 2 NRATE t + U t In which firms represents index of 50 distinguished companies in stock exchange, exchange shows currency rate in the free market, rate indicates real rate of five-year deposit bank interest, nexchange shows 185

5 A. H. Ahmadi & T. S. Emamgholi currency reference rate of Central Bank, nrate also indicates five-year deposit bank interest rate, α shows intercept, β indicates variable coefficient and U t shows random disruption term. In fact, the reason for using currency reference rate in the estimated model in addition to market currency rate and also utilizing nominal rate of bank besides the real bank interest rate is to estimate the effects of government policies in determining currency rate and bank interest on index of 50 distinguished companies. Before stating the estimate results, it should be mentioned that since all the variables of the model (dependent and independent variables) are at one side of the model in estimating long-term relationship with Johansen approach, for realizing effectiveness mark of each one of the independent variables on dependent variable, the plus mark should be interpreted as minus mark in estimate and minus mark as plus mark. The results of long-term relationship reveals that the real rate of five-year deposit bank interest and also currency rate in the free market have a significant negative effect on index of 50 distinguished companies in stock exchange, and as a result the first hypothesis stating that There is a significant relationship between stock price index in stock exchange and bank interest rate cannot be rejected, and the significant relationship of this hypothesis is accepted. According to the results displayed in table 3, currency rate in the free market has a significant negative effect on index of 50 distinguished companies in stock exchange. Concerning the results obtained from the table, the second hypothesis stating that There is a significant relationship between currency rate (market) fluctuations and stock price index in the stock exchange, which indicates a significant relationship between fluctuations of currency rate and index of the price of stock in stock exchange, is accepted and proved. But the significant result is that the nominal rate of deposit interest and currency reference rate have positive and significant effects on index of 50 distinguished companies in stock exchange. Therefore, it can be said that although currency rate in the free market and real bank interest rate are considered as an investment alternative for the capital market and have negative effects on index of 50 distinguished companies in stock exchange, concerning the significant effects of government policies on companies financial statements, they have caused currency reference rate and rate of bank interest have positive effects on 50 distinguished companies in stock exchange. Table 3: Estimating the effects of market rate and currency reference as well as real and nominal rate of bank interest variables on 50 distinguished companies in stock exchange through Johansen approach Vector Error Correction Estimates Date: 09/02/13 Time: 11:05 Sample (adjusted): 1387M M09 Included observations: 51 after adjustment Standard errors in () & t-statistics in [] CointEq1 Cointegrating Eq: FIRMS(-1) EXCHANGE (-1) ( ) [ ] RATE(-1) ( ) [ ] NEXCHANGE(-1) ( ) [ ] NRATE (-1) ( ) [ ] C

6 ANOVA of the Effects of Variables on 50 Distinguished Companies in Stock Exchange: ANOVA can help with examining share instability of variables in justification of specific fluctuations while instant reaction function illustrates the effect of shock to an endogenous variable on other variables of VAR model. ANOVA separates fluctuations in an endogenous variable comparing to shock to other endogenous variables; as a result, ANOVA provides information on the relative importance of each random shock for impressing variables of the model. In other words, ANOVA is used in order to examine share instability of variables in justification of specific fluctuations 5. The following table displays the results obtained from ANOVA of FEVDs for fluctuations of 50 distinguished companies in stock exchange comparing other effective variables. Table 4: ANOVA Model for 50 Distinguished Companies in Stock Exchange NRATE NEXCHANGE RATE EXCHANGE FIRMS S.E. Period The results obtained from ANOVA of 50 distinguished companies revealed that within 5 time periods, the most fluctuations in index of 50 distinguished companies in stock exchange among the variables under examination was due to real bank interest rate in a way that within 5 time periods, share of real bank interest rate is 18 %, market currency rate 8%and currency reference rate as well as nominal rate of bank deposit interest are less than 1% in all. But with increasing time periods, the role of market currency in fluctuation of index of 50 distinguished companies rises in a way that after 10, 15, and 20 time periods, 15, 40 and 49 percent of fluctuation of index of 50 distinguished companies in stock exchange will be respectively under the effect of market currency rate but share of the real bank interest rate will decrease during long timeperiods in a way that after 10, 15 and 20 time periods, the role of real bank interest rate will be 29, 18 and 13 percent respectively in fluctuations of index of 50 distinguished companies in stock exchange. It is worth mentioning that with increasing time intervals, the role of currency reference rate in fluctuations of index of 50 distinguished companies rises in a way that with increasing time intervals, the role and importance of currency reference rate is more than real bank interest rate though share of currency 5 Agrawalla, Raman K, (2010), Stock Market and the Real Economy: A policy perspective for India from Time Series Econometric Analysis presented at the 42nd Annual Conference of the Indian Econometric Society (TIES) PP

7 A. H. Ahmadi & T. S. Emamgholi reference rate is less than real bank interest rate during short periods (5 time periods) (although the role of market currency rate is more than currency reference rate in anyway) in a way that in time intervals of 10, 15 and 20, share of currency reference rate in fluctuations of index of 50 distinguished companies will be 10, 22 and 25 percent. Among other variables, nominal rate of bank deposit interest has the least role in fluctuations of index of 50 distinguished companies in stock exchange such that in time intervals of 10, 15 and 20, its share will be 7, 11, and 12 percent respectively. 5. Conclusion The results of the estimate revealed that nominal rate of bank interest and currency reference rate have positive and significant effects on index of 50 distinguished companies in stock exchange though real bank interest rate and market currency rate have negative effects on index of 50 distinguished companies in stock exchange. In fact, although market currency rate and real rate of bank deposit interest are considered as a competitor and alternative for capital market, government policies have positive effects on index of 50 distinguished companies and capital market in determining currency reference rate and bank deposit interest. As a result, research hypotheses stating the relationship between market currency rate and index of 50 distinguished companies in stock exchange and also the relationship between real bank interest rate and index of 50 distinguished companies in stock exchange are accepted. References Wang, Hung-Jen & Ho, Chia-Wen, "Estimating fixed-effect panel stochastic frontier models by model transformation," MPRA Paper 31081, University Library of Munich, Germany. Christopher Gan, David Cohen, Mike Clemes, Esther Chong, A Survey Of Customer Retention In The New Zealand Banking Industry, Banks and Bank Systems / Volume 1, Issue 4, Hesing, Y. (2011), Macroeconomic Determinats of the Stock Market Index and Policy Implication: The Case of Central European Country, Eurasian Journal of Business and Economics, 4 (7),PP Sarmad, Z.; Barzgan, A., Hejazi, Mehdi. (2011), Research methods in the behavioral sciences, Informed Publishing, pg 23. The Tehran Stock Exchange. (Various years) Monthly Statistical. Agrawalla, Raman K, (2010), Stock Market and the Real Economy: A policy perspective for India from Time Series Econometric Analysis presented at the 42nd Annual Conference of the Indian Econometric Society (TIES) PP Ashraf, M. A., and Noor, M. S. I. (2010) Impact of Capitalization, on asset Price Bubble in Dhaka Stock Exchange. Journal of Economic Cooperation and Development, 31(4), pp Bahmani -Oskooee, M. and A. Sohrabian, (2008), Stock Prices and the Effective Exchange Rate of the Dollar Applied Economics 24,pp Benderly, Jason,Zwich, Burton) 2009( Inflation Real Balances and Stock Returns American Economic Review, 2010, 15 pp Beirene, G M caporale and n Spagnolo (2009) Market, Interest Rate and Exchange Rate Risk Effects on Financial Stock Returns, Quantitative and Qualitative Analysis in Social Sciences Volume 3, issue 2, Graham, Fred, )2011( Inflation Real Stock Return And Monetray Policy, Applied Finanical Economics,2012, 6 pp

8 Chitnis, A. (2010) Performance Evaluation of Two Optimal Portfolios by Sharpe s Ratio. Global Journal of Finance and Management, ISSN , Vol. 2, No. 1, pp Kia, A. (2008). Forward Looking Agents and Macroeconomic Determinants of the Equity Price in a Small Open Economy. Applied Financial Economics, Vol. 13, PP Md. M Alam and Md. G S Uddin (2009). Relationship between Interest Rate and Stock Price: Empirical Evidence from Developed and Developing Countries, International Journal of business and Management Vol.4 PP