Integration of Foreign Exchange Markets: A Short Term Dynamics Analysis

Transcription

1 Global Journal of Management and Business Studies. ISSN Volume 3, Number 4 (2013), pp Research India Publications Integration of Foreign Exchange Markets: A Short Term Dynamics Analysis Amity Business School, Amity University, Haryana, India. Abstract The recent financial crisis in US and then debt crisis in Europe has impacted the economic performance of the world in general and financial performance in particular. These recent events have created a lot of volatility in the financial markets including currency markets. The volatility spillover depends upon the degree on integration between markets under consideration. This paper examines the integration of foreign exchange rates of Indian Rupees (INR), Great Britain Pound (GBP) and Euro (EUR) against US Dollar. Unit-root test was used to identify the stationarity of data; Granger Causality Test was used to identify the direction of causality. Impulse response was used to identify the effect of shocks from one market to other markets. Output suggests that INR Granger Causes GBP but GBP doesn t Granger Cause INR. INR Granger Causes EUR but EUR doesn t Granger Cause INR. EUR Granger Causes GBP but GBP doesn t Granger Cause EUR. The impulse response of GBP to INR suggests that any shocks from INR to GBP survive for a period of five days and that of from EUR to GBP survives for 3 days. Variance decomposition of returns of considered currency markets suggests that USD_INR is 100% influenced by own dynamics while other currency markets are well diversified and having less than 60% influenced by own dynamics. These outcomes suggest that Indian currency market is less integrated to foreign markets and hence, the volatility in Indian currency market is higher in comparison to other markets under consideration. This research will help portfolio managers and investors to take the advantage of diversification with a new perspective. Keywords: Granger Causality Test, Impulse Response, Integration, Unit Root Test.

2 Introduction With reduction in barriers of entry for foreign market participants in Indian markets, quantum of money flow has increased through both portfolio investments as well as direct investment. Indian corporate houses have borrowed huge amount of money from international markets in the form of FCCB (foreign currency convertible bonds) to take advantage of low cost of fund. Moreover, recent financial crisis in US and then debt crisis in Europe have deteriorated investor s sentiment and hence, volatility across all market segments has increased. For developing nations like India, an understanding of volatility spillover from foreign markets to Indian markets is of much importance as volatility in currency markets increases fluctuations in earnings of corporate (Kim 2003). Volatility spillover depends upon the degree of integration between markets and hence, integration of financial markets has attracted a number of researchers and practitioners. The study of financial market integration is important as it helps in formulating hedging strategies, security pricing, portfolio diversification and policy formulation. A portfolio manager tries to identify the degree of integration between different financial markets to diversify his portfolio (Sentana, 2000). The dynamics of financial markets are studied in two time frames: short term dynamics & long term dynamics. This paper examines the short-term dynamics of financial markets integration. Study of currency market integration has always been conducted taking developed markets as focus point (Palakkod 2012). This paper examines the integration of foreign exchange rates of Indian Rupees (INR), Great Britain Pound (GBP) and Euro (EUR) against US Dollar as well as evaluates the Transmission of price shocks from one market to other markets under consideration. Daily exchange rate data was taken from January 2008 to December 2012 for analysis from Google finance. 2. Methodology The data taken into consideration are time-series data and hence, Stationarity of data was checked using ADF (Augmented Dickey Fuller) test. ADF test is a well accepted test for unit root testing of time series data. If Yt is the time series to be tested for unitroot, then the test statistics for ADF unit root test is given by following relation: t n ρ yt 1 + μ + λt + αi yt i ut (1) i= 1 Δy = + The short-term dynamic relationships amongst markets were identified using Granger Causality Test. Granger Causality answers the question Does changes in one market cause change in other markets? Granger Causality test is sensitive to the lead lag relationship. Too many lags reduce the power of test due to estimation of additional parameters and a loss of degree of freedom. In contrast, too few lags may not capture the dynamics of the actual error correction process, resulting in poor estimation and its standard errors. To identify the appropriate lag length relationship, multivariate

3 Integration of Foreign Exchange Markets: A Short Term Dynamics Analysis 385 information criterion method was used, which suggested a lag order 2. Impulse Response methodology was used to trace out the responsiveness of the dependent market in the VAR (Vector Auto Regression) model to shocks to each of the markets. Variance Decomposition methodology was used to examine the proportion of the movements in the dependent market that are due to their own shocks, versus shocks to other markets. 3. Findings The summarized output of ADF test is as given below: Null hypothesis of Augment Dickey Fuller test is- Ho: Variable has a unit root. Case 1: Variable taken into consideration is the exchange rates. Variable t-statistics Prob. Conclusion USD_EUR Accepted USD_GBP Accepted USD_INR Accepted Case 2: Variable taken into consideration is return of respective exchange rates. Variable t-statistics Prob. Conclusion RUSD_EUR Rejected RUSD_GBP Rejected RUSD_INR Rejected The above output suggests that exchange rates data are non-stationary and returns on exchange rates are stationary data. Hence, returns data were taken into consideration for Granger Causality test as it requires stationary data. Null Hypothesis: Obs F-Statistic Probability RUSD_EUR does not Granger Cause RUSD_INR RUSD_INR does not Granger Cause RUSD_EUR RUSD_GBP does not Granger Cause RUSD_INR RUSD_INR does not Granger Cause RUSD_GBP RUSD_GBP does not Granger Cause RUSD_EUR RUSD_EUR does not Granger Cause RUSD_GBP Output suggests that RUSD_INR Granger Causes RUSD_GBP but RUSD_GBP doesn t Granger Cause RUSD_INR suggesting uni-directional relationship. RUSD_ INR Granger Causes RUSD_EUR as well as RUSD_EUR doesn t Granger Cause RUSD_INR providing no sign of relationship. RUSD_ EUR Granger Causes

4 386 RUSD_GBP but RUSD_GBP doesn t Granger Cause RUSD_EUR suggesting unidirectional relationship. Impulse response shown above represents the responsiveness of the dependent variable in the VAR to shocks to each of the variables. Granger Causality doesn t give any information about sign of relationship or how long these effects require to take place. Impulse response answers these questions. Output suggests that any shock from RUSD_INR to itself, RUSD_EUR survives for a period of 4 days and bear a positive relationship while any shock from RUSD_INR to RUSD_GBP bear a negative relationship and survives for 4 days. Similarly, any shock from RUSD_EUR to itself survive for a period of 2 days and bear a positive relationship while shock to RUSD_INR survives for a period of 4 days and bear a positive relationship. Any shock from RUSD_EUR to RUSD_GBP bears a negative relationship and survives for 4 days. Moreover, any shock from RUSD_GBP to itself survives for a period of 2 days in positive relationship and for next 2 days in negative relationship while shock to RUSD_INR survives for a period of 4 days and bear a positive relationship. Any shock from RUSD_GBP to RUSD_EUR bears a negative relationship and survives for 4 days. All impulse outputs suggest that there is volatility spillover between markets considered. Variance Decomposition of Period(Days) RUSD_INR RUSD_EUR RUSD_GBP RUSD_INR

5 Integration of Foreign Exchange Markets: A Short Term Dynamics Analysis 387 RUSD_EUR RUSD_GBP Variance decomposition provides the proportion of variance in the dependent variable that is due to their own shocks versus shocks to other variables. Outputs of variance decomposition also confirm the results of Granger Causality and Impulse response. RUSD_INR gets explained by itself only and hence, USD_INR market is does not get influenced by USD_EUR & USD_GBP markets. While, USD_INR & USD_GBP market explains 16.55% and 26.14% respectively the variance of USD_EUR market and hence, USD_EUR market gets influenced by USD_GBP more than that by USD_INR market. Moreover, USD_INR & USD_EUR influence USD_GBP market by 17.58% & 10.67%. Therefore, outputs of Granger Causality, Impulse Response & Variance Decomposition suggest that in short-term USD_INR market doesn t get influenced by USD_GBP & USD_EUR market while USD_EUR and USD_GBP markets get influenced by all three markets under consideration. 4. Conclusion The study of financial market integration is important as it helps in formulating hedging strategies, security pricing, portfolio diversification and policy formulation. A portfolio manager tries to identify the degree of integration between different financial markets to diversify his portfolio (Sentana, 2000). The integration dynamics of financial markets are studied in two time frames: short term dynamics & long term dynamics. This paper examines the short-term dynamics of financial markets integration. Granger Causality, Impulse response and Variance decomposition techniques were applied to study the short-term dynamics. Output suggests that INR Granger Causes GBP but GBP doesn t Granger Cause INR. INR Granger Causes EUR but EUR doesn t Granger Cause INR. EUR Granger Causes GBP but GBP doesn t Granger Cause EUR. The impulse response of GBP to INR suggests that any shocks from INR to GBP survive for a period of five days and that of from EUR to GBP survives for 3 days. Variance decomposition of returns of considered currency markets suggests that USD_INR is 100% influenced by own dynamics while other currency markets are well diversified and having less than 60% influenced by own dynamics. These outcomes suggest that Indian currency market is less integrated to foreign markets and hence, the volatility in Indian currency market is higher in comparison to other markets under consideration. This paper is focused on identification of integration between markets, explaining the reasons of the integration can be considered in future.

6 388 References [1] Brailsford, J., (1996), Volatility Spillovers across Tasman, Australian Journal of Management 21-1: [2] Bollerslev, T. (1990), Modelling the Coherence in Short-Run Nominal Exchange Rates: A Multivariate Generalized ARCH Approach, Review of Economics and Statistics 72, pp [3] Dickey, D.A. and W. Fuller (1981), Likelihood Ratio Tests for Autoregressive Time Series with a UnitRoot., Econometrica 49, pp [4] Kim, S. (2003) The Spillover Effects of US and Japanese Public Information News in Advanced Asia-Pacific Stock Markets. Pacific-Basin Finance Journal, 11(5): [5] Roll, R., (1989), Price Volatility, International Market Links, and Their Implications for Regulatory Policies, Journal of Financial Services Research 3, pp [6] Santana, M. H Novo Mercado, Focus No. 5. Washington, D.C.: Global Corporate Governance Forum.