Impact of Future Derivatives on Stock Market Volatility
Derivatives has been the talk of the financial world after it was accused as the primary reason for such a deep financial crisis that affected the global economy in 2007. Thus, the modelling of asset returns and judging the volatility of stock market and whether the derivatives have a substantial effect on stock market volatility, is still the key task for every finance professional as it provides much needed on risk patterns involved in investment process. The Finance Gurus propose that stock market normally exhibit high levels of price volatility and cause concerns to the investor regarding unpredictable outcomes, however with launch of derivatives in the nineties in the major financial hubs, the volatility of stock market has become more complicated with derivatives offering new areas and scope of hedging and speculation. Thus, it is important to examine the impact of derivatives on stock market volatility.
At time of introduction of derivatives, it was said that it is being launched with the twin objective of increasing liquidity and mitigation of risk. However, the finance world is still figuring it out whether these objectives have been materialized or not as it carries both theoretical and practical importance with it. Thus, with the help of stress testing models like GARCH and other models we will be looking on the impact of future derivatives on stock market volatility.
Introduction:
The completeness feature of the market has generated the need for introducing innovative financial instruments with the objective of creating efficient portfolios along with security from price fluctuations. Thus, with futures and other derivative instruments offering ability to benefit from upward or downward movement of prices, the demand for these financial instruments have been increasing over the years. However, with the introduction of derivatives in the financial world, more and more uninformed or irrational investors get attracted to stock market and with lower information received by the traders in the cash markets, stock market experience a higher volatility of the price fluctuation.(Alexander, 2001)
In order to study the impact of future derivatives we will be using the S&P CNX Nifty Index as a benchmark index and to account for non-constant error variance in the return series a GARCH Model will be used by incorporating futures and options dummy variables in the conditional variance equation.
However, before proceeding with our analysis we must look at what existing literature concludes about impact of future derivatives on stock market volatility:
RECENT LITERATURE:
A number of studies have been conducted so far and each study had come up with its own conclusion. While one set of analysis indicates that inclusion of derivatives does not lead to destabilization of underlying market rather it improves liquidity in the stock market. On the other hand, second set of analysts reveals that their study provide evidence that derivatives do increase the stock market volatility. Thus, the topic is still debatable in finance world but before proceeding with our analysis we must look what the recent literature concludes about impact of derivatives on stock market volatility:
- Rahman(2001) in order to estimate the conditional volatility and impact of futures trading on stock movement on Dow Jones Industrial Average Index(DJIA) used a GARCH model to estimate the movement in Intra day returns both before inclusion of futures and after inclusion of futures in the investor’s portfolio. He concluded that there is no change in volatility of stock returns in either period.
- Figuerola-Ferretti and Gilbert(2001) carried their analysis for the same purpose using auto regression model and and GARCH model. Along with these statistical model they also showed the results of VAR model to present the response analysisto track the effect of a shock to each of the volatilties. They concluded that inlcusion of derivatives futures decreases volatility.
- Bologna and Cavallo (2002) also studied the effect of future derivatives on Italian Markets to which on the basis of GARCH Model they concluded that the introduction of stock futures reduces the volatility of the index futures market.
- Chiang and Wang (2002) examined the impact of futures trading on Taiwan spot index volatility. Their study which was based on time varying GJR volatility model to which they showed that trading on futures on Taiwan Index stabilize the price volatility while MSCI Taiwan futures has no effects except asymmetric response behavior. Their study also discussed the macroeconomic and asymmetric effects of futures trading on spot price volatility behaviour.
- Thenmozi(2002) also conducted similar subject analysis on S& P Nifty Index of India because of recent introduction of Future Derivatives in Indian Stock Exchange. His study concluded that since introduction of futures trading the volatility of spot index return has reduced.
In contrast to above studies that concluded that introduction of future derivatives reduced the stock market volatility, following studies concluded that it actually increased the volatility of stock market:
- Lee and Ohk(1992) on their study relating to effects of future derivatives on stock market volatility on the financial markets of Japan, Hong Kong, UK, Australia nd USA. He concluded that except for the study of Australia and Hong Kong, derivatives indeed destabilized the markets and lead to stock prices volatility.
- Kamara(1992) also confirmed Lee and Ohkl view regarding US Market that futures trading destabilized the spot market by increasing the volatility of S&P 500 index of US Market.
- Yu(2001) who was the fist person in new millenium to use GARCH model for conducting the analysis on US, France, Japanese, Australia,Uk and Hong Kong markets that in contrast to Lee and Ohk view regarding Australian Financial Market, as he concluded that even the Austrian Stock Market got destabilized with future derivatives although he confirmed his view on Hong Kong Market that no volatility in stock prices were seen post-future trading.
Analysis: NIFTY Index
The Nifty Index:
The index was to trade on futures derivatives in June 2000. Both National Stock Exchange and Bombay Stock Exchange(BSE) trades in Derivatives but NSE accounts for 99.5% of the total derivative traded volume, thus our study will be focused on NSE Index of NIFTY. For the purpose of our analysis, we will be using daily closing prices of Spot Nifty Index, Nifty Index Futures, Nifty Junior Index and Spot S&P 500 Index from 5th October to 30th June 2006. Since NIFTY started trading in Futures from the year 2000 the prices were available from the day of initiation of futures in NIFTY. All the prices were collected mainly from official website of NSE and Yahoo Finance. Since our analysis required daily compounding return, we converted closing price into compunded returns through first log difference.
Data Series Source:
Of the total data derived from NSE and Yahoo Finance, we come up with daily returns of NIFTY index and NIFTY Junior Indices.
NIFTY Index:
With the total 2685 daily observations, the NIFTY Index provided a mean return of 0.0409% with standard deviation of 1.618%. Dividing our study into pre-futures and post-futures period where we assumed 12th June, 2000 to be the break off date, the average daily return during pre-futures period was 0.0294% and for the post-period it was 0.0497%. However, the interesting observation was that the standard deviation which is used here as measure of volatility decreases from 1.80% to 1.46% Standard deviation from pre-future introduction period to post-future introduction period.
Even if the break off date is changed a year later to June 2001, even then similar result was obtained as a decrease in spread with introduction of futures in the index.
NIFTY Junior Index:
While refering to daily prices and returns of NIFTY Junior Index, we come up with average daily mean return of 0.0523% and a standard deviation of 1.88%. Although, this index do not trade any derivative instrument over it, for the sake of indepth analysis of derivative futures on stock market performances, we break off the analysis period both pre-futures period and post-futures period. Before introduction of futures in the NIFTY Index, NIFTY Futures had a mean return of 0.0434% while the standard deviation was 2.034%. However, the results post futures introduction were different in comparison to NIFTY Index as now the mean daily returns reduced to 0.0434% and standard deviation reduced to 1.7538%. Thus, the return of NIFTY Junior Index was reduced during post futures period.
Intermediate Conclusion:
Our analysis revealed that stock which were part of NIFTY Index provided better returns after derivatives futures were introduced while stocks in NIFTY Junior Index on an average showed reduced returns and volatility post-futures were introduced. Overall on intermediate basis, we will conclude that introduction of derivatives indeed had an effect on stock market volatility. However, this conclusion may or may not change after we use GARCH model along with dummy variables to study the effect of derivative tradings.
Analysis Process:
In order to conduct our analysis over stock price volatility of NIFTY Index we will be using three methods to understand as what happens to the stock prices before and after the introduction of future derivatives. The fist method is the most simple method where volatility of stock prices is judged by the model of conditional volatility both for pre-futures and post-futures period. Then, going with advanced level testing we will go ahead with advanced level testing with GARCH Family models where our results will be confined with two popular GARCH Family models namely: GARCH(1,1) and EGARCH(1,1).
GARCH(1,1) Model
One of the most popular method among academicians, this method was developed by Boolersleva nd Taylor. GARCH modles have the ability to allow conditional variance to depend on its previous lags. Thus, this method allows interpretation of the current variance fitted in the model as weighted function of previous period volatility and fitted variance of previous period. Concluded to be better than ARCH Model, although GARCH(1,1) can be further generalized with GARCH(p,q) model but since GARCH(1,1) captures sufficient volatility most of the practitioners avoid GARCH(p,q) model.
GARCH/ARCH Effect
Testing for GARCH/ARCH effects
In the initial analysis, we performed GARCH/ARCH Model to study the their effects in the time series of returns of both NIFTY Index and NIFTY Junior Index. In order to study the effect it is important to use Lagrange Multiplier test in our model where we started with residual term in the mean equation for long four using the below given model:
The results of Lagrange Test are given below, followed by analysis:
**Notes: R2 = 0.1173, Adjusted R2 = 0.1159, Standard Error = 6.0879, Observations = 2681, F = 88.8807, Significance F = 0.
The above table indicates that regression test shows that coefficient for lag one intercept is significant at the level of 1%. The F Value calculated exceeds the LM test statistic value of 2.37 as TR2 is 2681*0.1173= 314.41, which does not follow the value of 9.4877 at 5% level of significance. Thus, hypothesis are rejected and it is proved that there are sufficient ARCH Effects.
Thus, since the residuals have sufficient ARCH/GARCH Effects, in order to conclude if derivative instruments affect stock market volatility, it is important to use suitable ARCH/GARCH Model to model volatility.Academicians suggest that GARCH(1,1), EGARCH and IGARCH will be suitable for Indian Stock Exchanges but since last two models are concerned with assymetric information and since we are least concerned with such provision, we will be using GARCH(1,1) model conditional variance model and to study the impact of derivatives instruments we will use following model:
In the above equation, VSi is the measure of volatility in the spot market during period t, Dt is the dummy variable assuming it attains value of 1 if ‘t’ is a post future time period after derivatives are included and zero for pre futures period of inclusion of derivatives while Bo,B1,B2 are the regression parameters. Dummy variable is important as it is included in the model with the assumption that it alone capture the direction of volatility of stock prices so that a positive significant value implies an increase in volatility while a negative significant value implies decrease in volatility with the introduction of futures. In other words, if value is not significant it means that derivative instruments do not affect stock market volatility.
Analysis of the model are displayed under:
The GARCH Analysis(With Dummy Variable)
Beginning with GARCH(1,1) model and in order to measure the more indepth impact of derivative instruments on stock market volatility we now introduce dummy variable to the GARCH(1,1) model. From the results provided with inclusion of dummy variable into GARCH(1,1) in the table below it is clearly evident that market wide factors concerned with Nifty Junior Returns, explain the return series of NIFTY Index while global factors concerned with lagged returns of S&P 500 Index do not indicate any information about returns of NIFTY Index. We have also included the day of week effect for all days of the week except Wednesday. Finally, the coefficients of futures dummy λ with coefficient of 0.504 and with t-ratio of 0.8955 is not significantly different from zero, which means that derivative instruments do not destabilize or stabilize the stock market volatility. Thus, our GARCH(1,1) model with dummy variable suggest that derivative instruments do not affect stock market volatility.
Final Conclusion:
The objective of our analysis was to conduct analysis as whether introduction of derivative instruments affects stock market volatility and for doing so we carried GARCH(1,1) model and GARCH(1,1) Model with inclusion of dummy variable. Both of the models suggest that introduction of derivatives in stock market does not destabilize or stabilize the stock market volatility. The seperate time period of pre-introduction of derivatives and post-introduction of derivatives reveals that the sensitivity of Nifty Index Returns to the Junior Nifty Index Returns and the day of the week effect disappears after derivatives are introduced. The sensitivity of prices to old news in the market is more influential during pre-futures period as compared to post-future period and once derivative instruments are introduced, stock market volatility is determined by innovations and also evidenced by higher ARCH Value. Large value of coefficients in ARCH and GARCH Model in the post introduction period of indicates that the stock returns are still influenced by past innovations and stock volatility is time varying. Also continuous shocks and long term memory process were observed in post derivatives introduction period, which makes us conclude that ‘’Introduction of derivatives do not affect stock market volatility’’ and any change in stock volatility is because of factors like better information and more transaprency in the market and not because of introduction of derivatives in stock trading.
References:
Afsal, Mallikarjunappa. THE IMPACT OF DERIVATIVES ON STOCK MARKET. Research Report. Kerela, India: ASIAN ACADEMY of Managment Journal of Accounting and Finance, 2008. PDF Document.
O'Connor. "Derivative Instruments." Institute, CFA. Derivatives. Boston: Custom, 2011. 11-34. Print.
Antoniou,Holmes. Futures trading, information and spot price volatility: Evidence for the FTSE-100 stock index futures contract using GARCH. Journal of Banking and Finance, 19, 117–129. 2010. Print
Antoniou, Pericili. Index futures and positive feedback trading: Evidence from major stock exchanges. Journal of Empirical Finance, 12, 219–238. 2005. Print
Bansal. An empirical re-examination of the impact of CBOE option initiation on the volatility and trading volume of the underlying equities: 1973–1986. The Financial Review, 24, 19–29. 1989. Print
Bauer. Forecasting volatility with encompassing and regime dependent GARCH models. Journal of Financial Management and Analysis, 18(2), 1–11. 2005. Print
Bollerslev. Generalized autoregressive conditional heteroscedasticity. Journal of Econometrics, 31, 307–327. 1986. Print
Bologna, Covallo. Does the introduction of stock index futures effectively reduce stock market volatility? Is the 'futures effect' immediate? evidence from the Italian Stock Exchange using GARCH. Applied Financial Economics, 12, 183–192. 2002. Print
Boyer, Popiela. Index futures and stock price volatility.Derivatives Use, Trading & Regulation, 9(4), 351–364. 2004. Print
Calado, Pareira. An empirical analysis of the effects of options and futures listing on the underlying stock return volatility: The Portuguese case. Applied Financial Economics, 15, 907–913. 2005. Print
Chamberlain,Cheung and Kwan. The impact of options listing on stock behavior and market liquidity: some Canadian evidence. Journal of Business Finance and Accounting, 20, 687–698. 2010. Print
Chiang, Wang. The impact of futures trading on spot index volatility: evidence for Taiwan index futures. Applied Economics Letters, 9, 381– 385. 2002. Print
Conrad. The price effect of options introduction. Journal of Finance, XLIV, 487–98. 1989. Print
Drimbetas, Porfiris. The effect of derivatives trading on volatility of the underlying asset: Evidence from the Greek Stock Market. Applied Financial Economics, 17(2), 139–148. 2007. Print
Engle. Autoregressive conditional heteroskedasticity with estimates of the variance of the United Kingdom inflation. Econometrica, 50, 987–1007. 1998. Print
