Creating Value in Small-cap Firms by Mitigating Risks of Market Volatility

Abstract

Extant research has explored numerous ideal approaches to predict and anticipate the unpredictability in stocks to mitigate business risks. This article attempts to offer an important insight on creating values in terms of financial returns dodging the risks associated with the market volatility in emerging market economies by exploring the context of National Stock Exchange (NSE), India. The study establishes that Small-cap companies, which are included in NSE Small 100 index, are less inclined to be impacted by the market volatility index (NVIX) compared to the Large-cap companies and Mid-cap companies that are under respective Broad Market Indices. Furthermore, this article examines 64 Small-cap companies, belonging to nine different sectors, to investigate the sector-wise impact of market volatility on Small-cap businesses in India.

Keywords

Small Cap Firms Risk Market Volatility National Stock Exchange

Introduction

For the prospect of high returns, financial specialists look for prominent emerging markets, primarily because they are more likely to achieve faster economic growth as measured by growth rate of GDP than that of the developed economies (Caprio & Honohan, 1999; Rousseau & Sylla, 2003). However, the investments in emerging markets often accompany serious uncertainty because of political instability, infrastructure problems, currency fluctuation and regulations imposed by government authorities (Clemente, 1994; Schnabl & Hoffmann, 2008; Dhir, 2017; Dhir & Dhir, 2015; Dhir & Mital, 2012). This article attempts to provide an important insight to the investors on creating values in terms of financial returns dodging the risks associated with the market volatility.

Volatility alludes to the measure of instability or the span of changes in a security’s worth (Edwards, 1988; Mariana & Semmler, 2002; Morck, Yeung & Yu, 2000). A higher volatility implies that a stock’s worth can possibly be spread out over a large span (Illing & Liu, 2006; Marianne & van Dijk, 2004). This implies the share prices can change drastically over a brief span on either side (Bushee & Christopher, 2000; Edwards, 1988; Haugen, Talmor & Torous, 1991). Volatility indices, often called as fear indices, provided by different stock exchanges worldwide are used as prevalent measures of the implied volatility in the concerned markets (Andersen & Bollersley, 1998; Cumby, Figlewski & Hasbrouck, 1993). These volatility indices usually adopt the methodology of calculating VIX index—which Chicago Board Options Exchange uses for S&P 500 stock index option prices—for concerned stock markets (Brenner & Galai, 1989; Dhaene et al., 2012; Guillaume & Schoutens, 2012). The National Stock Exchange (NSE) in India provides daily data of NVIX, which is widely accepted as Indian volatility index (Kumar & Jaiswal, 2013).

This study attempts to address two major objectives. First, to compare the impact of NVIX on the Broad Market Indices designed for Large-cap, Mid-cap and Small-cap companies. Second, the impact of NVIX on the Small-cap companies listed in NSE. Post introduction, this article is divided into five broad sections. In the first section, discussion of the literature related to the topic of this article leads to the development of hypothesis. The second section reveals the sample formation and time period of this study. The third section describes the methods of data preparation and data analysis that has been followed to test the hypothesis in detail. The fourth section analyzes the findings and proposes implications for managers and practitioners. The final section is devoted to discuss the limitation and future research scopes along with implications for researchers regarding this study.

Theoretical Background

A lot of research has been devoted to find out the best way to model, and thereafter predict, the volatility in stock markets (Bagchi, 2014; Wang & Moore, 2009). Some of the important studies in the late 1990s and early 2000s (such as Fleming, Ostdiek & Whaley, 1995; Whaley, 2000) established the argument that implied volatility contains all relevant information to explain the realized volatility. During that time, Moraux, Navatte and Villa (1999) found a strong relation between future realized volatility in French market and volatility index (VX1). Also, Poon and Granger (2003) had observed that implied volatility outperform the other competing volatility forecasts (such as historical returns volatility, lagged realized volatility, and ARCH\GARCH conditional volatility).

Yang and Liu’s (2012) empirical evidences show that the volatility index (TVIX) is a strong indicator of Taiwanese stock market volatility, and TVIX outperforms the stock index returns volatility forecasts (e.g., historical and GARCH). Siriopoulos and Fassas (2012) concluded that Greek implied volatility index (GRIV) best explains the future realized volatility in Greek stock market beyond that impound in the historical volatility. Similarly, the Indian volatility index NVIX will be the most appropriate benchmark to study the impact of market volatility on Indian companies listed in NSE, India. Thus, we hypothesize:

H1: Impact of NVIX on NSE Small 100 index (SC100) is lesser as compared to Nifty 50 index, Nifty next 50 index, and Nifty Mid-cap 100 index.

Extant research (Glosten & Milgrom, 1985; Roll, 1984) shows that at least one source of volatility can be explained by the liquidity provision process. When the stock market infers the possibility of adverse selection, they adjust their trading ranges, which, in turn, increase the band of price oscillation, and hence generate volatility (Chakravarty, Gulen & Mayhew, 2004; Easley & O’Hara, 1992; Sandås, 2001). Hence, to investigate the outcome regarding our first hypothesis on individual Small-cap companies, we can develop a second hypothesis that comprises two parts as follows:

H2A: The share prices of Small-cap companies listed in NSE are more volatile than NVIX.

H2B: The share prices of Small-cap companies listed in NSE are more volatile than NIFTY Small-cap 100 index.

Sample and Data

The scope of this study is restricted to investigate 100 companies that comprise NSE Small-cap 100 index. Moreover, nine different sectors are identified for which standard NIFTY Sectoral Indices are available. These nine sectors include Automobiles, Banks, Energy, Financial Services, Infrastructure and Realty, IT, Media, Metals and Pharmaceuticals. Out of 100 companies under consideration, only 64 belong to these nine sectors and are included in this study (Table 1). The data of this study range from 2 March 2009, the day NSE introduced NVIX, to 31 December 2015, the last day of latest completed quarter, that is, Q3 of FY2015–2016. Adjusted closing prices of the shares of all 64 companies along with NVIX and NIFTY Small-cap 100 index and Sectoral Indices for 1696 trading days during the specified 82-month window are collected from the official website of NSE India. For those companies, which were included in the NIFTY Small-cap 100 index after 2 March 2009, time series of NVIX and NIFTY Small-cap 100 Index indices are truncated accordingly during data analysis.

Methodology

Depending on the size of market capitalization, stocks are divided into three categories—Large-cap, Mid-cap and Small-cap stocks. For example, companies with market capitalization in between INR 10 billion and INR 50 billion as on 1 January 2003 were included in the list of NIFTY Mid-cap 100 stocks [then, 1 USD = 45 INR, approximately]. Two popular Broad Market Indices for Mid-cap and Small-cap companies, namely Nifty Mid-cap 100 Index and Nifty Small-cap 100 Index, respectively, represents about 13.86 and 3.03 per cent, respectively, of the free-float market capitalization of all stocks listed on NSE as on 31 March 2015. Whereas Nifty 50 Index, the flagship index of NSE, and Nifty Next 50 Index mostly include Large-cap companies diversified into 13 sectors, and together represent about 78.19 per cent of the free-float market capitalization in NSE as on the last day of FY2014–2015.

Data Preparation

The time series data of NVIX is found to be non-stationary with a pattern of random walk with drift and stochastic trend. Hence, the data are made stationary in two steps (Pouzols & Lendasse, 2010). First, percentage changes in closing price of NVIX were calculated for obtaining first difference. Second, de-trending was done by subtracting the arithmetic mean of the percentage changes from each value of percentage change that is obtained in the previous step. Equation (1) provides the stationary data series for NVIX. Similarly, data was prepared for the four Broad Market Indices, nine Sectoral Indices, and 64 companies as discussed above.

Table 1.

Sixty-four Companies in the Sample

Sectors (with Sectoral Indices)	No. of Companies in the Respective Sector	Listed Names of the Companies	Relative Risk (w.r.t NVIX)					Relative Risk (w.r.t SC100)
Sectors (with Sectoral Indices)	No. of Companies in the Respective Sector	Listed Names of the Companies	Very Low	Low	Medium	High	Very High	Very Low	Low	Medium	High	Very High
AUTOMOBILE (Nifty Auto)	4	AMTEKAUTO, CEATLTD, ESCORTS, JKTYRE	0	2	2	0	0	0	3	1	0	0
BANK (Nifty Bank)	15	ALBK, ANDHRABANK, CENTRALBK, CORPBANK, DENABANK, INDIANB, IOB, KTKBANK, LAKSHVILAS, ORIENTBANK, SBBJ, SBT, SYNDIBANK, UCOBANK, VIJAYABANK	0	1	13	1	0	0	0	15	0	0
ENERGY (Nifty Energy)	5	ABAN, CHENNPETRO, JPPOWER, PTC, SJVN	0	1	2	2	0	0	1	2	2	0
FIN-SERV (Nifty Financial Services)	8	IFCI, JMFINANCIL, MAGMA, MUTHOOTFIN, PFS, RELIGARE, SEINV, SREINFRA	0	4	0	4	0	0	3	1	4	0
INFRA-STRUCTURE & REALTY (Nifty Realty)	13	INDIACEM, PRISMCEM, DELTACORP, GMRINFRA, GVKPIL, GODREJPROP, HCC, IBREALEST, KEC, NBCC, PUNJLLOYD, SOBHA, UNITECH	3	2	3	4	1	3	3	3	3	1
IT (Nifty IT)	8	FSL, HEXAWARE, KPIT, NIITTECH, POLARIS, ROLTA, TATAELXSI, ECLERX	0	1	2	4	1	0	1	2	4	1
MEDIA (Nifty Media)	3	DEN, EROSMEDIA, TV18BRDCST	0	1	1	0	1	0	0	2	0	1
METALS (Nifty Metal)	4	GMDCLTD, JINDALSAW, MOIL, ORISSAMINE	0	0	1	3	0	0	0	1	3	0
PHARMA (Nifty Pharma)	4	ASTRAZEN, FDC, MARKSANS, SUVEN	0	0	1	1	2	0	0	1	1	2

Source: Authors’ own.

Note: Detailed data and results are available on request.

\begin{matrix} N V I X_{t} = [{(p e r c e n t c h a n g e i n c l o \sin g p r i c e o f N V I X)}_{t} - \\ (m e a n o f t h e p e r c e n t c h a n g e s o f N V I X)] \end{matrix}

(1)

Variables

To fulfil the first objective of our study, four Broad Market Indices, namely NSE SC100, Nifty 50 index, Nifty next 50 index and Nifty Mid-cap 100 index, are one by one taken as dependent variables against NVIX as the independent variable in all four cases. Next, to fulfil the second objective of our study, the closing prices of all 64 companies in our sample are taken as dependent variables against the corresponding sectoral indices as independent variables along with either NVIX (for H2A) or NSE Small 100 index—SC100 (for H2B).

Data Analysis

To test the first hypothesis, four Broad Market Indices, as mentioned before, are regressed by NVIX using ordinary least squares (OLS) technique. The summarized results in Table 2 show that there is no spurious relationship present in any case, but goodness of fit is found to be very low in the fourth case where Nifty Small-cap 100 index was taken as dependent variable. To test the second hypothesis, Logistic Regression technique was used. The co-efficients a_k and b_k of the independent variables NVIX and Sectoral Indices, respectively, for the first Logit model were estimated according to Equation (2) using ‘Eviews Software’ at 95 per cent confidence level. For example, the first out of 64 regression equations is given by Equation (3). It was observed that all R² values are less than Durbin–Watson statistics. Also, no significant problem of co-integration occurred in any of the 64 cases. Similarly, to estimate co-efficients for the second Logit model, Equation (4) is used to estimate co-efficients p_k and q_k of the independent variables SC100, denoting NSE SC100, and Sectoral Indices, respectively, at 5 per cent significance level, for the second Logit model. In this model also, neither any spurious result nor the problem of cointegration is found in any of the 64 cases.

Table 2.

Regression Results

Dependent Variable	Slope (Co-efficient)	p-Value	Std. Error	t-stat.	R² Value	DW Stat.	AIC
NSE Nifty 50	−0.114	0.000	0.005	−23.245	0.242	1.775	2.999
NSE Next 50	−0.111	0.000	0.005	−21.179	0.209	1.684	3.142
NSE Mid-cap 100	−0.107	0.000	0.005	−22.025	0.223	1.621	2.991
NSE Small-cap 100	−0.042	0.000	0.006	−7.051	0.029	1.671	3.396

Source: Authors’ own.

\begin{matrix} {(D e p e n d e n t V a r i a b l e)}_{k, t} = c + a_{k} * N V I X_{t} + b_{k} * \\ {(S e c t o r i a l I n d e x)}_{t} + e_{t} \end{matrix}

(2)

\begin{matrix} {(A M T E K A U T O)}_{k, t} = c + a_{1} * N V I X_{t} + b_{1} * \\ {(N I F T Y ​ A U T O)}_{t} + e_{t} \end{matrix}

(3)

\begin{matrix} {(D e p e n d e n t V a r i a b l e)}_{k, t} = c + p_{k} * S C 100_{t} + q_{k} * \\ {(S e c t o r a l I n d e x)}_{t} + e_{t} \end{matrix}

(4)

Logit Regression

The first Logit model is given by the Equation (5), where Y_k,t is a binary variable where value ‘1’ represents the corresponding company that exhibits more volatility than NVIX and value ‘0’ represents otherwise. Z_k,t values are calculated using the value of estimated co-efficients. P_k,t, which denotes the probability of Y_k,t = 1, can be obtained by the expression derived in Equation (6) (Tabachnick & Fidell, 2007). Using the P_k,t value, we can calculate the Relative Risk given by the Odds Ratio in Equation (7). The odds ratio signifies the chance of occurring Y = 1 over chance of occurring Y = 0. Hence, an odds ratio of greater than unity value signifies Y = 1 is more likely than Y = 0.

\begin{matrix} Z_{k, t} = Y_{k, t} f o r {(D e p e n d e n t V a r i a b l e)}_{k, t} = c + a_{k} * \\ N V I X_{t} + b_{k} * {(S e c t o r i a l I n d e x)}_{t} + e_{t} \end{matrix}

(5)

\begin{matrix} Z_{k, t} = Y_{k, t} f o r {(D e p e n d e n t V a r i a b l e)}_{k, t} = c + a_{k} * \\ \begin{array}{l} N V I X_{t} + b_{k} * {(S e c t o r i a l I n d e x)}_{t} + e_{t} \\ L o g_{e} [(P_{k, t} / (1 - P_{k, t})] = Z_{k, t} \end{array} \end{matrix}

(6)

O d d s R a t i o = P_{k, t} / (1 - P_{k, t})

(7)

Result and Analysis

Table 2 shows strong support for the first hypothesis as the slope of NSE SC100 is found to be much flatter than that of Nifty 50 index, Nifty next 50 index and Nifty Mid-cap 100 index with respect to NVIX.

On the other hand, strong support, with Odds Ratio more than 1.02, for H2A and H2B is found only in 24 and 22 out of 64 cases, respectively. So, second hypothesis is not supported overall. Table 1 summarizes the decisions that are based on values of Odds Ratio, where relative risks are labelled as ‘Very High’, ‘High’, ‘Medium’, ‘Low’ and ‘Very Low’ subject to the discriminating values 1.1, 1.025, 0.975 and 0.9 of the Odds Ratio. The findings related to H2 are found to be fairly consistent with both of the measures used for H2A and H2B.

Managerial Implications

The discussions in the previous section indicate that Small-cap companies are less likely to be affected by the market volatility and, hence, involve lesser risk compared to overall market condition. This insight may help the investors while designing their portfolio. More specifically, among the nine sectors in our study, four sectors—Financial Services, IT, Metal and Pharmaceuticals—exhibited higher affinity towards volatility than the other five sectors–namely Automobiles, Banks, Energy, Financial Services, Infrastructure and Realty and Media.

Conclusion and Discussion

As shown in the Table 2, low R²-value in the case where Nifty SC100 was regressed by NVIX index may question the applicability of NVIX as a measure for the relationship between market volatility and Small-cap companies. This argument is backed by the fact that Nifty Small-cap 100 Index represents only about 3 per cent of the free-float market capitalization in NSE. Therefore, this article proposes a requirement of a standard and universally accepted volatility index for the Small-cap companies.

This article comes with many limitations and scopes of improvement including, but not limited to, the following issues:

The availability of NVIX data restricted the time period of this study to past seven years only. Similar researches in other stock markets, especially from other emerging economies, may both relax this limitation and investigate the universality of the findings of the study.

The scope of this study is confined to the NSE SC100 companies only. Hence, the impact of market volatility on Small-cap companies in India could be explored more extensively with a larger number of Indian companies, listed in NSE and/or Bombay Stock Exchange (BSE).

This article only analyzes the volatility or the risk associated with listed Small-cap companies in India. The relationship between risk and return for the companies in our sample was not explored.

This empirical study suggests that the Small-cap companies are less likely to be affected by the market volatility. Possible reasons behind this argument can be explored. In addressing to the last limitation mentioned above, research can be dedicated to explore the impact of globalization on the Small-cap companies from emerging economies with limited global exposure against that of similar companies with considerably higher global exposure. Further, the companies in the sample are divided into nine sectors irrespective of revenue and debt–equity structure of the companies. Similar to the sectoral analysis that has been performed in this study, the impact of market volatility on the Small-cap companies can be explored on the basis of aforesaid parameters. Finally, the volatility in share prices of different Small-cap companies can also depend on industry life cycle, technology life cycle, and company life cycle. Hence, different contingents of companies affiliated to different stages of those three life cycles may be analyzed to obtain valuable insight on this issue. Even after paying due importance to the limitations of this study, it can be concluded that the outcome of this article stands significant. This article argues in favour of the investment strategy to include the stocks of the Small-cap companies in a portfolio to mitigate risks, especially in the context of emerging markets. However, while executing the mentioned strategy, sectorial analysis seeks careful attention as discussed in this article.

References

Andersen

Torben G.

, & Bollerslev

(1998). Answering the skeptics: Yes, standard volatility models do provide accurate forecasts. International Economic Review, 39(4), 885–905.

Bagchi

(2014). Application of the concept of dissonance to explain the phenomenon of return-volatility relationship. IUP Journal of Applied Finance, 20(2), 5–17.

Brenner

Menachem

, & Galai

(1989). New financial instruments for hedging changes in volatility. Financial Analysts Journal, 45(4), 63.

Bushee Brian

, & Noe Christopher

(2000). Accounting information and the economics of the firm. Journal of Accounting Research, 38(1), 171–202.

Caprio

, & Honohan

(1999). Restoring banking stability: Beyond supervised capital requirements. The Journal of Economic Perspectives, 13(4), 43–64.

Chakravarty

, Gulen

, & Mayhew

(2004). Informed trading in stock and option markets. The Journal of Finance, 59(3), 1235–1257.

Clemente

(1994). Investing in Asia’s emerging equity markets. The Columbia Journal of World Business, 29(2), 92–111.

Cumby

, Figlewski

, & Hasbrouck

(1993). Forecasting volatility and correlations with EGARCH models. Journal of Derivatives, 1(2), 51–63.

Dhaene

, Dony

, Forys

M. B.

, Linders

, & Schoutens

(2012). FIX: The fear index—measuring market fear. Topics in numerical methods for finance (Vol. 19, pp. 37–55). Boston, MA: Springer Proceedings in Mathematics & Statistics.

10.

Dhir

(2017). Flexibility in modification and termination of cross-border joint ventures. Global Journal of Flexible Systems Management, 18(2), 139–151.

11.

Dhir

, & Dhir

(2015). Diversification: Literature review and issues. Strategic Change, 24(6), 569–588.

12.

Dhir

, & Mital

(2012). Decision-making for mergers and acquisitions: The role of agency issues and behavioral biases. Strategic Change, 21(1/2), 59–69.

13.

Easley

, & O’Hara

(1992). Adverse selection and large trade volume: The implications for market efficiency. Journal of Financial and Quantitative Analysis, 27(2), 185–208.

14.

Edwards

F. R.

(1988). Does futures trading increase stock market volatility? Financial Analysts Journal, 44(1), 63.

15.

Fleming

, Ostdiek

, & Whaley

R. E.

(1995). Predicting stock market volatility: A new measure. Journal of Future Markets, 15(3), 265–302.

16.

Glosten

L. R.

, & Milgrom

P. R.

(1985). Bid, ask and transaction prices in a specialist market with heterogeneously informed traders. Journal of Financial Economics, 14(1), 71–100.

17.

Guillaume

, & Schoutens

(2012). Calibration risk: Illustrating the impact of calibration risk under the Heston model. Review of Derivatives Research, 15(1), 57–79.

18.

Haugen

R. A.

, Talmor

, & Torous

W. A.

(1991). The effect of volatility changes on the level of stock prices and subsequent expected returns. The Journal of Finance, 46(3), 985–1007.

19.

Illing

, & Liu

(2006). Measuring financial stress in a developed country: An application to Canada. Journal of Financial Stability, 2(3), 243–265.

20.

Kumar

A. V.

, & Jaiswal

(2013). The information content of alternate implied volatility models: Case of Indian markets. Journal of Emerging Market Finance, 12(3), 293–321.

21.

Mariana

, & Semmler

(2002). Market share instability and stock price volatility during the industry life cycle: The US automobile industry. In Economic evolution, learning, and complexity (pp. 97–126). Berlin and Heidelberg: Springer-Verlag.

22.

Marianne Sensier

, & Van Dijk

(2004). Testing for volatility changes in U.S. macroeconomic time series. The Review of Economics and Statistics, 86(3), 833–839.

23.

Morck

, Yeung

, & Yu

(2000). The information content of stock markets: Why do emerging markets have synchronous stock price movements? Journal of Financial Economics, 58(1), 215–260.

24.

Moraux

, Navatte

, & Villa

(1999). The predictive power of the French market volatility index: A multi horizons study. European Finance Review, 2(3), 303–320.

25.

Poon

Ser-Huang

, & Granger

Clive W. J.

(2003). Forecasting volatility in financial markets: A review. Journal of Economic Literature, 41(2), 478–539.

26.

Pouzols

F. M.

& Lendasse

(2010). Effect of different detrending approaches on computational intelligence models of time series. Barcelona: IEEE World Congress on Computational Intelligence.

27.

Roll

(1984). A simple implicit measure of the effective bid-ask spread in an efficient market. Journal of Finance, 39(4), 1127–1139.

28.

Rousseau

P. L.

, & Sylla

(2003). Financial systems, economic growth, and globalization (Globalization in Historical Perspective). National Bureau of Economic Research. Chicago: University of Chicago Press.

29.

Sandås

(2001). Adverse selection and competitive market making: Empirical evidence from a limit order market. The Review of Financial Studies, 14(3), 705–734.

30.

Siriopoulos

, & Fassas

(2012). The efficiency of VIX futures market—A panel data approach. Journal of Alternative Investments, 14(3), 55–65.

31.

Schnabl

, & Hoffmann

(2008). Monetary policy, vagabonding liquidity and bursting bubbles in new and emerging markets: An overinvestment view. The World Economy, 31(9), 1226–1252.

32.

Tabachnick

B. G.

, & Fidell

L. S.

(2007). Using multivariate statistics. Boston: Pearson/Allyn & Bacon.

33.

Wang

, & Moore

(2009). Sudden changes in volatility: The case of five Central European stock markets. Journal of International Financial Markets, Institutions and Money, 19(1), 33–46.

34.

Whaley

R. E.

(2000). The investor fear gauges. Journal of Portfolio Management, 26(3), 12–17.

35.

Yang

M. J.

, & Liu

M. Y.

(2012). The forecasting power of the volatility index in emerging markets: Evidence from the Taiwan stock market. International Journal Economics and Finance, 4(2), 217–231.