Impact of Derivative Market on Stock Market Volatility
Impact of Derivative Market on Stock Market Volatility
This paper studies the volatility implications of the introduction of derivatives on stock market volatility in India using the S&P CNX Nifty Index as a benchmark. To account for non-constant error variance in the return series, a GARCH model is fitted by incorporating futures and options dummy variables in the conditional variance equation. We find clustering and persistence of volatility before and after derivatives, while listing seems to have no stabilisation or destabilisation effects on market volatility. The postderivatives period shows that the sensitivity of the index returns to market returns and any day-of-the-week effects have disappeared. That is, the nature of the volatility patterns has altered during the post-derivatives period. Keywords: conditional volatility, heteroscedasticity, volatility clustering, market efficiency
INTRODUCTION The modelling of asset returns volatility continues to be one of the key areas of financial research as it provides substantial information on the risk patterns involved in investment and transaction processes. A number of works have been undertaken in this area. Given the fact that stock markets normally exhibit high levels of price volatility, which lead to unpredictable outcomes, it is important to examine the dynamics of volatility. With the introduction of derivatives in the equity markets in the late nineties in the major world markets, the volatility behaviour of the stock market has become further complicated as derivatives open new avenues for hedging and speculation. The derivatives market was launched mainly with the twin objectives to transfer risk and to increase liquidity, 43
T. Mallikarjunappa and Afsal E. M.
thereby ensuring better market efficiency. The examination of how far these objectives have materialised is important both theoretically and practically. In India, trading in derivatives started in June 2000 with the launch of futures contracts in the BSE Sensex and the S&P CNX Nifty Index on the Bombay Stock Exchange (BSE) and National Stock Exchange (NSE), respectively. Options trading commenced in June 2001 in the Indian market. Since then, the futures and options (F&O) segment has been growing continuously in terms of new products, contracts, traded volume and value. At present, the NSE has established itself as the market leader in this segment in India, with more than 99.5 percent market share (NSE Fact Book, 2006, p. 85). The F&O segment of the NSE outperformed the cash market segment with an average daily turnover of Rs291.91 billion, as compared to Rs114.79 billion in the cash segment from 2006 to 2007 (Derivatives Updates on NSE website, www.nseindia.com, 2007). This shows the importance of derivatives in the capital market sector of the economy.
Previous studies on the volatility effects of derivatives listing provide mixed results, suggesting case-based biases. In addition, in India, there is a lack of robust examination of the impact of derivatives on market volatility. In India, trading in derivatives contracts has existed for the last six years, which is an adequate time period to evaluate its major pros and cons. Against this backdrop, it is important to empirically examine the impact of derivatives on the stock market. In this paper, we attempt to study the volatility implications of the introduction of derivatives on the cash market. Through this study, we seek evidence regarding whether the listing of futures and options lead to any significant change in the volatility of the cash market in India. In contrast to a sectoral index studied in previous research from Mallikarjunappa and Afsal (2007), we select a general index called the S&P CNX Nifty Index to which the first derivatives contract was introduced by the NSE in India.
The previous study noted the peculiar characteristics of IT stocks and arrived at the conclusion that stock-specific characteristics must be studied for any general conclusion. As a benchmark index, the Nifty Index is expected to show wider, more balanced and more applicable results and thus can be treated as a true replica of the Indian derivatives market. Most of the Indian studies, such as Thenmozhi (2002), Sibani and Uma (2007) and Mallikarjunappa and Afsal (2007), did not consider options contract, but this study examines the introduction of options while also analysing volatility. The period under analysis spans from October 1995 through June 2006. Furthermore, to allow for a non-constant error variance in the return series, we applied a GARCH model that was more appropriate to describe the data collected. Therefore, the present work offers a valuable addition to the existing literature and should prove to be useful to investors as well as regulators, as this is a broader index than the one studied by Mallikarjunappa and Afsal (2007).
The Impact of Derivatives on Stock Market Volatility
The remainder of this paper is organised as follows. Recent literature is briefly reviewed in Part 2, and Part 3 presents the econometric model, data and methodology. The empirical results of our work are discussed in Part 4, and Part 5 presents our conclusion. RECENT LITERATURE Various studies on the effects of futures and options listings on the volatility of an underlying cash market have been carried out across the world. Overall, the empirical evidence is mixed, and most studies suggest that the introduction of derivatives does not destabilise the underlying market. These studies also show that the introduction of derivatives contracts improves liquidity and reduces informational asymmetries in the market. However, some evidence exists in support of increased volatility with the onset of derivatives trading. Thus, the volatility implications of derivatives are still debatable. In this section, we consider the important and recent literature in this area. Rahman (2001) examined the impact of index futures trading on the volatility of component stocks for the Dow Jones Industrial Average (DJIA).
The study used a simple GARCH (1, 1) model to estimate the conditional volatility of intra-day returns. The empirical results confirm that there is no change in conditional volatility from pre- to post-futures periods. Figuerola-Ferretti and Gilbert (2001) used error-correction models and the GARCH (1, 1) regression model to study the effect of futures trading on volatility. In addition, they reported the results of a VAR model and presented an impulse response analysis to track the effects of a shock to each of the volatilities. The results show that volatility decreases in the post-futures period. Bologna and Cavallo (2002) examined the effect of the introduction of stock index futures for the Italian market. Their empirical results show that the introduction of stock index futures affects the volatility of the spot market.
In addition, the results from various GARCH (1, 1) models for pre-futures and post-futures sub-periods suggest that the index futures market reduces volatility. Chiang and Wang (2002) examined the impact of futures trading on Taiwan spot index volatility. Their study also discussed the macroeconomic and asymmetric effects of futures trading on spot price volatility behaviour. They used an asymmetric time-varying GJR volatility model. Their empirical results showed that the trading of futures on the Taiwan Index has stabilising impacts on spot price volatility, while the trading of MSCI Taiwan futures has no effects, except asymmetric response behaviour. Thenmozhi (2002) examined whether there was any change in the volatility of the S&P CNX Nifty Index in India due to the introduction of Nifty futures and whether movements in futures prices provided predictive information regarding subsequent movements in index prices.
The study shows that the inception of futures trading has reduced the volatility of spot index returns. Pilar and Rafael (2002) analysed the effect of the introduction of derivatives on the Ibex-35 Index using a dummy variable and a GJR model to test the impact of the introduction of derivative markets on the conditional volatility of the underlying asset. They found that although the asymmetry coefficient increased, the conditional volatility of the underlying index declined after derivatives were introduced. Robert and Michael (2002) investigated the impact of the introduction of stock index futures trading on the seasonality of daily returns of the underlying index for seven national markets. The results indicate reduced seasonality with respect to mean returns, thus leading to more efficiency in these markets. Shembagaraman (2003) explored the impact of the introduction of derivative trading on cash market volatility using data on stock index futures and options contracts traded on the Nifty Index.
The results suggest that futures and options trading has not led to a change in the volatility of the underlying stock index, but the nature of volatility seems to have changed in the post-futures market. The study also examined whether greater futures trading activity in terms of volume and open interest was associated with greater spot market volatility. It found no evidence of any link between trading activity variables in the futures market and spot market volatility. Sung, Taek and Park (2004) studied the effect of the introduction of index futures trading in the Korean markets on spot price volatility and market efficiency of the underlying KOSPI 200 stocks relative to the carefully matched non-KOSPI 200 stocks; they found evidence that market volatility was not affected by futures trading, while market efficiency was improved.
Taylor (2004) tried to uncover the determinants of trading intensity in futures markets. In particular, the time between adjacent transactions on the FTSE 100 index futures market was modelled using various augmentations of the basic autoregressive conditional duration (ACD). As predicted by various market microstructure theories, he found that the bid-ask spread and transaction volume have a significant impact on subsequent trading intensity. However, there was evidence that a large (small) difference between the market price and the theoretical price of the futures contract, which is known as pricing error, leads to high (low) levels of trading intensity in the subsequent period.
Boyer and Popiela (2004) looked into whether the introduction of futures to the S&P500 Index altered the effect of addition to, or removal from, the S&P500 Index. This study used the S&P500 price effect to show that overall price volatility did not show any significant increase for added stocks after trading began on the S&P500 Index futures. Calado, Garcia and Pereira (2005) used data for eight derivative products to study the volatility effect of the initial exchange listing of options and futures on the Portuguese capital market. They did not find significant differences in the unadjusted and adjusted variance and beta for the underlying stocks after the listing of derivatives. However, some of the underlying stocks taken individually have experienced significant increases or decreases in variance after derivatives listing.
Finally, they concluded that the introduction of a derivatives market in the Portuguese case has not had the average stabilisation effect on risk as detected in other markets. Gannon (2005) tested contemporaneous transmission effects across volatilities of the Hong Kong stock and index futures markets and futures volume of trade by employing a structural systems approach. Competing measures of volatility spillover, constructed from the overnight S&P500 Index futures, were tested and found to impact asset return volatility and volume of trade patterns in Hong Kong. Antoniou, Koutmos and Pericli (2005) tested the hypothesis that the introduction of index futures has increased positive feedback trading on the spot markets of six industrialised nations. Their findings support the view that futures markets help stabilise underlying spot markets by reducing the impact of feedback traders and attracting a greater number of rational investors.
Floros and Vougas (2006) examined the effect of futures trading on the volatility of the underlying spot market taking the FTSE/ASE-20 and FTSE/ASE Mid 40 Indices in Greece. The results for the FTSE/ASE-20 Index suggest that futures trading has led to decreased stock market volatility, but the results for the FTSE/ASE Mid 40 Index indicate that the introduction of stock index futures has led to increased volatility, while the estimations of the unconditional variances indicate a lower market volatility after the introduction of stock index futures. Sibani and Uma (2007) used OLS and GARCH techniques to capture the time-varying nature of volatility and volatility clustering phenomenon of the Nifty Index due to the introduction of futures trading.
The results suggest that there are no significant changes in the volatility of the spot market of the Nifty Index, but the structure of volatility changes to some extent. The study also reported that new information is assimilated into prices more rapidly than before, and there is a decline in the persistence of volatility since the introduction of futures trading. Drimbetas, Nikolaos and Porfiris (2007) explored the effects of the introduction of futures and options into the FTSE/ASE 20 Index on the volatility of the underlying index using an EGARCH model. It is shown that the introduction of derivatives induces a reduction of conditional volatility in the FTSE/ASE20 Index and consequently increases its efficiency. Mallikarjunappa and Afsal (2007) studied the volatility behaviour of the Indian market by focusing on the CNX IT Index, which is a sectoral index, and found that underlying volatility increases with the onset of futures trading. Their result contradict many other studies carried out in India, and it is reasoned that the sectoral index showed different behaviour in terms of returns and volatility, especially during the 2001 period of market scam in India.
They attributed these results to a sharp decrease in the prices of IT stocks after the stock market scam broke out in 2001. Since the sectoral index showed different results than those of earlier studies, these results must be examined as to whether they hold for the Indian market when a broader market index is studied. Their study also pointed out that results depend on the time period as well as the country studied. These results indicate the needed scope for further research as well as suggest the relevance of different samples and methodologies. DATA AND METHODOLOGY Data As reported in the introduction section, in India, futures trading on the S&P CNX Nifty Index of the NSE and the BSE Sensex Index of the BSE started in June 2000. NSE accounts for about 99.5 percent of the total trading volume in the derivatives segment; therefore, we use the S&P CNX Nifty Index to study the volatility behaviour of the market.
This study uses the daily closing prices of the Spot Nifty Index, the Nifty Index Futures, the Nifty Junior Index and the spot S&P500 Index from October 5, 1995, through June 30, 2006. For the Nifty futures, the data from June 12, 2000, onwards is used as the futures trading commenced from this day. The S&P CNX Nifty spot and futures and the Nifty Junior Index price data were collected from the NSE website (www.nseindia.com). The S&P500 Index price series was collected from Yahoo! Finance (www.yahoofinance.com).
The closing price data were converted to daily compounded returns by taking the first log difference. Return Rt at time t is given by Rt = ln( Pt / Pt −1 ) *100 , where Pt is the closing price for day t. The S&P CNX Nifty Index is a well-diversified index of 50 stocks comprising 25 sectors of the Indian economy. The average total traded value for the six months ending in June 2006 for all Nifty stocks was approximately 49.8 percent of the traded value of all stocks on the NSE. The Nifty stocks represent about 56.5 percent of total market capitalisation as of March 31, 2006. The next most liquid security after the S&P CNX Nifty Index is the CNX Nifty Junior Index. The maintenance of the S&P CNX Nifty and the CNX Nifty Junior Indices are synchronised so that the two will always be disjoint sets; that is, a stock will never appear in both indices at the same time. The CNX Nifty Junior Index represents about 9.77 percent of the total market capitalisation as of March 31, 2006.
The S&P500 Index is an index consisting of 500 stocks. It is one of the most commonly-used benchmarks for overall US equities and is meant to reflect the risk/return characteristics of large-cap stocks. Econometrics Techniques Assuming a constant error variance throughout the time period (that is, a homoscedasticity model), volatility measures like estimated standard deviation, rolling standard deviation and so on were developed to study the behaviour of stock market volatility (Hodgson & Nicholls, 1991; Herbst & Meberly, 1990). These studies implicitly assume that price changes in spot markets are serially uncorrelated and homoscedastic. However, the findings on heteroscedasticity in stock returns are well documented (Mandelbrot, 1963; Fama, 1965; Bollerslev, 1986; Shembagaraman, 2003; Nath, 2003).
In the presence of heteroscedasticity, the usual OLS estimates do not render the best linear unbiased estimator (BLUE) (Gujarati, 2005, p. 387). Stock market returns assume conditional and unconditional variances; the former relates to contemporaneous or short-term shocks and is unlikely to be constant over time. The latter is assumed to be constant. Thus, the disturbance or error term in the stock return series normally exhibits ‘varying’ variance and hence requires heteroscedasticity as a treatment.
In a seminal work, Engle (1982) proposed the Auto Regressive Conditional Heteroscedasticity (ARCH) process to model conditional variance. In an ARCH framework, the error variance is a function of the squared error variance in the previous term. To avoid the long lag lengths on the disturbance term, Bollerslev (1986) suggested the generalised ARCH, known as GARCH (p, q), in which the lags of the variance terms are also included in the variance equation. In this model, q refers to the lag on ε t2−1 (that is, the squared disturbance term), and p refers to the lag on ht (that is, the variance).
University/College: University of Chicago
Type of paper: Thesis/Dissertation Chapter
Date: 4 January 2017
We will write a custom essay sample on Impact of Derivative Market on Stock Market Volatility
for only $16.38 $12.9/page