Sentimental Herding in Stock Market: Evidence from India

Abstract

The present study examines herd behaviour in the Indian equity market using daily prices of 11 broad market indices, 12 sectoral indices, 16 strategic indices and 11 thematic indices from 1 January 2016 to 31 March 2022. For examining herd behaviour, daily returns, overnight returns, intraday spread, absolute market return and squared market return are calculated. Nifty 500 is used as a proxy index for the whole market. The regression models suggested by Christie and Huang (1995) and Chang et al. (2000) are used to calculate Cross-Sectional Standard Deviation (CSSD) and Cross-Sectional Absolute Deviation (CSAD) respectively, which measure the equity return dispersions. It is observed that herding exists in the Indian equity market when we calculate return dispersions using intraday spread. However, no herd presence is found when return dispersions are calculated through daily return, overnight return, absolute market return and squared market return. These results may be useful for investors, policy makers and regulatory bodies. Investors might design their investment strategy while making investment decision. Regulators might be able to formulate their policies to ensure market efficiency in the Indian equity market.

Keywords

Herd Behaviour Market Sentiment Behavioural Finance Return Dispersions

Introduction

The asset pricing model assumes that investors are rational and they tend to behave in such a manner so as to maximize returns. Markets are assumed to be efficient and all information is reflected in the stock prices, thereby making the investors behave rationally (Fama, 1970). However, investors’ sentiments come in between, which oppose the above hypothesis (Shiller, 2003). Investors’ feelings, emotions and psychology play a significant part in the choice to invest. Financial researchers have long been perplexed by investor attitudes regarding stock market trading (Kanojia et al., 2020a). The asset pricing model has been chastised for failing to forecast financial decisions (Kahneman & Tversky, 1979). This happens because of the unrealistic assumption of rationality (Rashid et al., 2019). The investors’ irrationality leads to the paradigm of behavioural finance (Kumar et al., 2016), which is linked with understanding the reasoning pattern of investors and affects their investment decision making (Mertzanis & Allam, 2018).

The present article focuses on one of the aspects of behavioural finance, that is, herd behaviour, which means imitating the actions of others, ignoring their own rational thinking and arriving at market consensus (Bikhchandani & Sharma, 2000). It is doing the same what every other person is doing even if private information does not favour it (Banerjee, 1992) and people responding in accordance with the majority of others around them, characterized by a lack of independent judgement or deliberation (Kanojia et al., 2020b). Individuals encounter comparable difficulties, which lead to similar options and as a result, similar payoffs (Bikhchandani et al., 1998). Investors feel optimistic while following the trend and rely on market sentiment for their investment decisions (Choi & Yoon, 2020). Herding opposes the rational asset pricing model.

Literature (Banerjee, 1992; Bikhchandani et al., 1992; Chevalier & Ellison, 1999; Hong et al., 2000; Scharfstein & Stein, 1990) suggests certain theories about herding. First, psychology effect leads to herding since sense of security is achieved by following the majority (Goldbaum, 2008). Second, it has been observed that informed investors would make a better investment decision rather than the uninformed ones, therefore, favouring information driven effect (Calvo & Mendoza, 2000). Reputation Effect constitutes the third theory, which suggests that mutual fund managers are trusted on account of their trading strategies, which builds their reputation (Swank & Visser, 2008). Investors do follow the herd in the same direction as they want some confirmation, which helps them to gain confidence during the periods of uncertainty (Nofsinger & Sias, 1999). Herding occurs because of various reasons, but the most obvious one is conformity with the society (West, 1988).

Herding is categorized into two parts: One is the spurious herding in which investors react to the information set and take the decisions following the same trend and second is the intentional herding where investors imitate others because they aim to do so (Bikhchandani & Sharma, 2000). The financial perspective of herding categorizes it into two dimensions: irrational herding in which the investors, not agreeing by this ideology, ignore their own analysis and rely on the market consensus (Christie & Huang, 1995). While rational herding is executed to protect one’s own reputation. This situation is compared with the situation of traders and agents whose performance is evaluated on their performance. In order to achieve good performance, they are likely to depend on others who have varied knowledge in this regard.

It is observed that investors tend to collect each and every information prior to making stock market investment (Mertzanis & Allam, 2018). In contrast to it, during extreme conditions, such activities of collecting information and then taking investment decisions accordingly, is quite impossible (Dhall & Singh, 2020). At times of stress, people start following others, which leads to inefficiency in the market (Nofsinger & Sias, 1999). Therefore, markets start destabilizing and prices tend to diverge from their fundamental value leading to low returns. Hence, biased earnings come into the picture, which weakens the financial market structure, making the entire financial system look fragile (Kumar et al., 2016).

A strand of literature, whether stock prices herd around the market, is available. Some studies (see, Abd-Alla, 2020; Chen, 2013; Choi & Yoon, 2020; Dhall & Singh, 2020; Economou, 2019; Luu & Luong, 2020; Poshakwale & Mandal, 2014; Vo & Phan, 2019) observed that during extreme conditions of market stress, individuals are more likely to follow the market consensus and tend to ignore their own beliefs. However, other studies (Chong et al., 2017; Garg & Gulati, 2013; Kumar et al., 2016; Yousaf et al., 2018) witnessed contradicting patterns of herd behaviour. The mixed evidence relating to herd presence is found with respect to time and market conditions.

Luu and Luong (2020) did sectoral analysis for the period 2000–2020 and examined the existence of herd behaviour in Taiwan and Vietnam stock markets during the pandemic influenza and COVID-19. Herd presence was found in insurance and materials in Vietnam and in real estate in Taiwan. Herding was noticed in the transportation industry in Vietnam and the energy industry in Taiwan during the influenza pandemic. Similarly, COVID-19 outbreak led to no herd behaviour in Vietnam while herding persists in the food and beverage industry of Taiwan. Furthermore, Choi and Yoon (2020) found herd presence in down market periods, high-trading volume periods and low and high quantiles in KOSPI and KOSDAQ stock markets for the period January 2003 to December 2018. The analysis also took into account the times of extreme crisis. Hence, herding formation was noticed in transportation industry in Vietnam and energy industry in Taiwan during Influenza pandemic. COVID-19 outbreak led to no herd behaviour in Vietnam while herding persists in Food Beverage Industry of Taiwan. Economou (2019) concluded that a glimpse of herding was witnessed in Romania. Bulgaria also witnessed herding during the low market volatility days for the period October 2000–December 2016. Vo and Phan (2019) conducted industry-wise analysis for the period 2005–2016 and observed the existence of herding in the banking sector of financial industry at times of rising market. However, weak evidence of herding was found in technological industry. Chen (2013) studied the herd behaviour with respect to developing, emerging and frontier markets for the period 1 January 2000–31 December 2009. The frontier markets witnessed the highest average daily return, followed by the emerging and developed markets. It was observed that herd behaviour persists in both up and down markets and also the tendency of investors to herd was more in response to bad news rather than good news.

Chang et al. (2000) tested the herd behaviour on account of investment behaviour seen in various international markets. The results indicated that equity return dispersions for the US, Hong Kong and Japan showed an increase, therefore, rejecting the presence of herd behaviour in these markets. While, the results regarding South Korea and Taiwan witnessed the presence of herd behaviour because they showed smaller equity return dispersions.

However, Christie and Huang (1995) examined the herd presence in equity returns. It was found that monthly data showed greater dispersions. Beta values from daily regression equations were positive, and thereby, strongly favoured the rational asset pricing model. Yousaf et al. (2018) stated that herd behaviour did not exist during extreme market movements as beta coefficients are positive. However, herding existed during low volume trade days and during crisis period, that is, 2007–2008. Chong et al. (2017), in their sample period for the period January 2000 to December 2011, found that coefficients of absolute market return were significantly positive in up and down markets, thereby rejecting the presence of herd behaviour.

While, in the Indian context, Kanojia et al. (2020b) observed no significant evidence of herding in the Indian stock market during different market conditions for the period 1 April 2009– 31 March 2018. Dhall and Singh (2020) could not capture herd presence in the Indian equity market for the period 1 January 2015–1 June 2020. However, when sectoral analysis was done, then instances of herd behaviour were found in auto industry, IT and pharma sectors in the post-COVID-19 times during bullish market conditions. Furthermore, during bearish market conditions, industry-wise analysis posit same results. Herd presence was found in media and entertainment industry. As opposed to it, Kumar et al. (2016) found no significant evidence of herd behaviour in the Indian stock market for the whole research period, including both up and down markets and dramatic price swings for the period 1 January 2008–31 December 2015. Garg and Gulati (2013) also witnessed increased equity dispersions during extreme market movements, thereby, rejecting any herd presence in the Indian stock market for the period April 2000–April 2013.

This study aims to contribute in terms of herd existence in up and down markets and during extreme periods of stress. It is beneficial to study the existence of herd behaviour in the Indian equity market as it would help the investors in predicting the future movement of the stocks correctly so that stocks could be valued properly and they might earn good returns. Both academicians and policy makers would get an insight from the study.

Hence, in the backdrop of the earlier discussion, this study is an attempt to contribute to the earlier yet unsettled debate by examining the herd behaviour in the Indian stock market using four categories of indices of the National Stock Exchange of India (NSE), namely broad market indices, sectoral indices, strategic indices and thematic indices from 1 January 2016–31 March 2022. This sample period has witnessed phenomenal growth in terms of trading volume, making it as a manifest period chosen for the testing of herd existence in India. To the best of our knowledge, this is the maiden study, which uses the data of all categories of indices of equity segment at the NSE. The study covers all categories of indices representing the equity segment at NSE as these serve as benchmark index for examining the performance of the stocks. It may also be noted that the previous studies (Abd-Alla, 2020; Chen, 2013; Choi & Yoon, 2020; Dhall & Singh, 2020; Economou, 2019; Garg & Gulati, 2013; Kanojia et al., 2020b; Kumar et al., 2016; Luu & Luong, 2020; Poshakwale & Mandal, 2014; Vo & Phan, 2019; Yousaf et al., 2018) have considered daily prices of individual stocks, while this article is considering stock market indices as a whole to examine whether it makes any difference in the findings and implications of herd behaviour in India. Therefore, the current study aims to fill this research gap and examine the herd existence in Indian equity market.

The rest of the article is organized as follows. The second section explains data description. The third section discusses research methodology, the fourth section provides results with interpretation, the fifth section concludes the study and discusses the future scope of the study.

Data Description

Data for this study has been extracted from the website of the NSE of India (www.nseindia.com) for four categories of indices namely broad market indices, sectoral indices, strategic indices and thematic indices (Table 1). The sample period for all indices is from 1 January 2016–31 March 2022. The rationale behind taking this sample period is the phenomenal growth in the trading volume since 2016. Figure 1 shows that trading quantity and turnover at NSE up to the period 2015–2016 is almost constant, however, it gained momentum in 2016 and witnessed escalation during 2019–2020. This is also evident from Figure 2, which shows that there is a notable rise in average daily turnover and number of trades during this period. Since trading activity is an important factor in the literature of market efficiency, therefore, the study of this sample period may provide important insights.

Table 1.

List of Broad Market Indices, Sectoral Indices, Strategic Indices and Thematic Indices.

Broad Market Indices	Sectoral Indices	Strategic Indices	Thematic Indices
1. NIFTY 50	1. NIFTY Auto	1. NIFTY 50 Dividend Points	1. NIFTY Commodities
2. NIFTY 100	2. NIFTY Bank	2. NIFTY 50 Equal Weight	2. NIFTY CPSE
3. NIFTY 200	3. NIFTY Financial Services	3. NIFTY 50 PR 1X Inverse	3. NIFTY Energy
4. NIFTY Mid Cap 50	4. NIFTY Financial Services 25/50	4. NIFTY 50 PR 2X Leverage	4. NIFTY India Consumption
5. NIFTY Mid Cap 100	5. NIFTY FMCG	5. NIFTY 50 TR 1X Inverse	5. NIFTY INFRA
6. NIFTY Mid Cap 150	6. NIFTY IT	6. NIFTY 50 TR 1X Leverage	6. NIFTY Mid Cap Liquid 15
7. NIFTY Mid Small Cap 400	7. NIFTY Media	7. NIFTY 50 Value 20	7. NIFTY MNC
8. NIFTY Next 50	8. NIFTY Metal	8. NIFTY 100 Equal Weight	8. NIFTY PSE
9. NIFTY Small Cap 50	9. NIFTY Pharma	9. NIFTY 100 Low Volatility 30	9. NIFTY Services Sector
10. NIFTY Small Cap 100	10. NIFTY Private Bank	10. NIFTY 100 Quality 30	10. NIFTY 100 ESG Sector Leaders
11. NIFTY Small Cap 250	11. NIFTY PSU Bank	11. NIFTY 200 Momentum 30	11. NIFTY 100 LIQ 15
	12. NIFTY Realty	12. NIFTY 200 Quality 30
		13. NIFTY Alpha 50
		14. NIFTY Alpha Low Volatility 30
		15. NIFTY Dividend Opportunities 50
		16. NIFTY Growth Sectors 15

Source: www.nseindia.com

Figure 1.

Source: Based on data extracted from the official website of NSE under the head ‘Business Growth Segment in Capital Market’.

Figure 2.

Source: Based on data extracted from the official website of NSE under the head ‘Business Growth Segment in Capital Market’.

Moreover, the Nifty 500 is used as a proxy index as it is a comprehensive index of 500 firms that account for 90% of India’s total market capitalization, making it a useful proxy index. Additionally, it represents 98% of total turnover of trades in India (Katarachia et al., 2018).

Research Methodology

This study is focused to examine the herding behaviour with the help of equity returns. In this study, five types of returns are calculated.

Daily Return: Daily return on stock is calculated as:

R_{t} = \frac{(P_{t} - P_{t - 1})}{P_{t - 1}}

(1)

where P_t = closing price of stock at time t.

Overnight Return: Overnight return is calculated as:

Overnight {Return}_{t} = \frac{({Open}_{t} - {Close}_{t - 1})}{{Close}_{t - 1}}

(2)

where Open_t = opening price of stock at time t.

Close_t−1 = closing price of stock at time t−1.

Intraday Return: Intraday return is calculated as:

Intraday {Return}_{t} = \ln \frac{{High}_{t}}{{Low}_{t}}

(3)

where High_t = high price of stock at time t.

Low_t = closing price of stock at time t.

Absolute Market Return: Absolute market return is calculated as the absolute value of daily stock returns.

{Absolute Market Return}_{t} = ABS (\frac{P_{t} - P_{t - 1}}{P_{t - 1}})

(4)

Squared Market Return: Squared market return is calculated as the square of daily stock returns.

Squared M a r k e t {Return}_{t} = {(\frac{P_{t} - P_{t - 1}}{P_{t - 1}})}^{2}

(5)

To examine the herd behaviour in security prices, regression models suggested by Christie and Huang (1995) and Chang et al. (2000) are used. The first model tests whether the dispersions are lower during the times of extreme market conditions. It was argued that individual investors tend to ignore their own information, rely on others and behave accordingly in the stock market. As a result, the security returns will not deviate too far from the market and investors will invest keeping in mind the collective action of the market. In order to measure the proximity of individual returns to the market average, cross-sectional standard deviation is calculated with respect to the market return.

{CSSD}_{t} = \sqrt{\frac{\sum_{i = 1}^{N} {(R_{i t} - R_{m t})}^{2}}{N - 1}}

(6)

where CSSDt = Cross-Sectional Standard Deviation on day t; R_it = stock returns i at time t; R_mt = market return at time t; N = number of stocks.

Dispersions must be low to validate the herd presence. However, low dispersions do not guarantee the herd presence. Sometimes, lack of new information prevailing in the market leads to low dispersions and herding is not witnessed.

According to the asset pricing model assuming rationality, periods of market stress lead to increase in dispersions. In contrast, when herding exists, dispersions are lower. In order to test this, the level of dispersions is isolated in the extreme tails of market distribution to see whether the security returns are significantly different from the market return. This testing is done using the following regression model:

C S S D = α + β_{1} U p + β_{2} D o w n + ε

(7)

where CSSD is the Cross-Sectional Standard Deviation. ‘α’ coefficient denotes the average dispersions in the sample. ‘Up’ and ‘Down’ are two dummy variables. ‘Up’ refers to the dummy variable, which is equal to 1 when the daily market returns lie in the extreme upper tail of observation and otherwise 0. Similarly, ‘Down’ refers to the dummy variable, which is equal to 1 when the daily market returns lie in the extreme lower tail of the observation and otherwise 0.

This study uses extreme 5% criteria for the upper and lower tail of market returns distribution. The herd presence is validated when the β₁ and β₂ coefficients are significantly negative while positive coefficients of β₁ and β₂ will disconfirm any herd presence. Positive coefficients of β₁ and β₂ will indicate the high volatility among individual stock returns.

There are limitations to CSSD model. The CSSD model examines herding with respect to the extreme market conditions only and it is overshadowed due to the presence of outliers. These limitations are taken into account in CSAD model. CSAD model is proposed by Chang et al. (2000). This model extended the work of Christie and Huang (1995) and is less exposed to outliers. This model examines the non-linear relationship between equity return dispersions and the overall market return and uses CSAD instead of CSSD, which is defined as:

{CSAD}_{t} \frac{\sum_{i = 1}^{N} | R_{i t} - R_{m t} |}{N}

(8)

where CSADt = Cross-Sectional Absolute Deviation on day t; R_it = stock returns i at time t; R_mt = market return at time t; N = number of stocks.

It is observed that in the presence of herding, increase in market return leads to decrease in individual equity returns dispersions. This is in contradiction to the asset pricing model, which states that there is a linear relationship between equity return dispersions and market return. This linear relation will no longer hold good, when the market participants tend to follow the herd and imitate market behaviour. In order to check the validity, one more variable r²_m,t become the part of the regression equation. The herding behaviour is examined using the following regression model:

C S A D = α + γ_{1} r_{m},_{t} + γ_{2} | r_{m},_{t} | + γ_{3} r^{2}_{m},_{t} + ε

(9)

where CSAD is the Cross-Sectional Absolute Deviation. r_m,_t is the cross-section average return of sample on day t. |r_m,t| is the absolute market return at day t. r²_m,t is the squared value of the market return on day t.

A significantly negative γ₃ coefficient implies the herd presence. This indicates the herd presence during extreme stress period and observes a non-linear relation between return dispersions and squared market return, thereby stating that investors typically ignore their own convictions and do what the market dictates while taking investment decisions. A significantly positive γ₃ coefficient will indicate that market return will increase with the increase in CSAD, hence, giving evidence against herding.

Results and Interpretation

This study examines herding with the help of daily return, overnight return, intraday return, absolute market return and squared market return using the regression models suggested by Christie and Huang (1995) and Chang et al. (2000). Cross-Sectional Standard Deviation (CSSD) and Cross-Sectional Absolute Deviation (CSAD) are calculated using the above models. The lower the dispersions, the higher would be the chances of herd existence. Descriptive statistics are calculated with regard to cross-sectional standard deviation and cross-sectional absolute deviation using broad market indices, sectoral indices, strategic indices and thematic indices.

Table 2 provides the descriptive statistics of CSSD and CSAD for broad market indices, sectoral, strategic indices and thematic indices. The high difference between the maximum and minimum values of CSSD and CSAD indicates higher volatility in CSSD and CSAD. The value of kurtosis is more than 3, which indicates that the data does not follow the normal distribution. The higher values of skewness would suggest that the outliers form the part of dataset.

Table 2.

Descriptive Statistics (CSSD and CSAD).

		Close Prices		Overnight Return		Intraday Return		Absolute Market Return		Squared Market Return
		CSSD	CSAD	CSSD	CSAD	CSSD	CSAD	CSSD	CSAD	CSSD	CSAD
Broad market indices	Mean	0.005903	0.004047	0.003184	0.001988	0.005687	0.004319	0.005464	0.003694	0.002677	0.000872
	Median	0.003989	0.002884	0.001694	0.001271	0.004649	0.003497	0.003622	0.002589	4.50E-05	3.24E-05
	Maximum	1.114247	0.321186	1.113511	0.320423	0.041962	0.033078	1.114247	0.321186	3.751297	1.131066
	Minimum	0.000612	0.000443	0.000136	9.77E-05	0.000864	0.000628	0.000571	0.000385	9.97E-07	7.14E-07
	Standard deviation	0.029494	0.009922	0.029487	0.009767	0.004028	0.003194	0.029499	0.009887	0.09553	0.028821
	Skewness	34.91331	25.75788	35.16285	27.65155	2.427274	2.430959	34.93964	26.08723	39.12576	39.05084
	Kurtosis	1295.826	759.1903	1306.768	830.8685	13.73811	13.34268	1297.041	773.1339	1535.615	1531.597
Sectoral indices	Mean	0.010664	0.007845	0.003662	0.002686	0.011963	0.008887	0.00945	0.006747	0.000392	0.000233
	Median	0.009455	0.006928	0.00281	0.002041	0.010862	0.008167	0.008266	0.005845	0.000204	0.000115
	Maximum	0.200218	0.191215	0.203423	0.194724	0.067866	0.037257	0.199909	0.190943	0.044463	0.042563
	Minimum	0.001211	0.000951	0.000794	0.000601	0.002653	0.002165	0.001211	0.000951	1.15E-05	6.18E-06
	Standard deviation	0.007349	0.006049	0.005736	0.005218	0.005359	0.003696	0.007288	0.005956	0.001423	0.00115
	Skewness	12.63945	18.5913	27.72588	32.41407	2.781099	1.984317	13.15376	19.72149	23.67573	32.73001
	Kurtosis	297.6334	548.5805	954.119	1188.213	19.86537	10.15713	313.847	594.3771	670.8668	1192.123
Strategic indices	Mean	0.008464	0.007896	0.004759	0.003546	0.007764	0.005628	0.005645	0.005234	0.000214	0.000152
	Median	0.006416	0.006161	0.003172	0.002462	0.006416	0.004692	0.004425	0.004138	6.34E-05	5.12E-05
	Maximum	0.20508	0.102912	0.204634	0.1965	0.097453	0.056957	0.201603	0.055683	0.044273	0.014253
	Minimum	0.00102	0.000701	0.000391	0.000285	0.00184	0.001319	0.000815	0.000562	2.46E-06	7.71E-07
	Standard deviation	0.009184	0.006867	0.007595	0.006274	0.005839	0.003881	0.006915	0.004374	0.001343	0.000532
	Skewness	9.49916	4.723868	14.1941	20.15695	5.881098	4.713396	16.56259	3.58803	25.83783	15.77769
	Kurtosis	163.3279	42.76247	328.8589	587.4242	62.04497	40.25977	430.898	25.85778	789.4737	352.9347
Thematic indices	Mean	0.006575	0.004976	0.002661	0.001975	0.006957	0.005227	0.005734	0.004297	0.00018	0.000129
	Median	0.005731	0.004307	0.002023	0.001513	0.006142	0.004651	0.004913	0.00367	8.14E-05	5.38E-05
	Maximum	0.202187	0.192646	0.204304	0.194789	0.059331	0.036565	0.202187	0.192646	0.044826	0.04274
	Minimum	0.001588	0.001025	0.000524	0.000396	0.00109	0.000813	0.000867	0.000662	3.90E-06	2.32E-06
	Standard deviation	0.005966	0.005366	0.005525	0.005086	0.003811	0.002699	0.005879	0.005318	0.001173	0.001106
	Skewness	23.15977	27.86604	31.71576	35.33479	3.715411	2.970902	24.48226	28.93948	35.88434	37.15366
	Kurtosis	749.5132	969.1955	1149.982	1337.631	33.56479	22.08595	808.6221	1019.754	1360.715	1429.259

Source: The authors.

Note:

• CSSD stands for cross sectional standard deviation.

• CSAD stands for cross sectional absolute deviation.

Furthermore, to examine the existence of herd behaviour first, CSSD regression is calculated by using the Christie and Huang (1995) model and results are presented in Table 3. In this model, beta coefficients are calculated. Significantly negative beta coefficients support herding in the stock market while positive beta coefficients reject the herd presence in the stock market. It is observed that in spite of significant market swings, herd behaviour is not observed in the Indian stock market.

Table 3.

CSSD Regression.

	Daily Return		Overnight Return		Intraday Return		Absolute Market Return		Squared Market Return
	UP	DOWN	UP	DOWN	UP	DOWN	UP	DOWN	UP	DOWN
Broad market indices	0.004877 (0.15)	0.003042 (0.37)	0.003736 (0.27)	0.003508 (0.30)	0.006992* (0.00)	−0.002070* (0.00)	0.007328* (0.03)	−0.001467 (0.67)	−0.00174 (0.88)	−0.00288 (0.80)
Sectoral indices	0.009445* (0.00)	0.007353* (0.00)	0.007305* (0.00)	0.005117* (0.00)	0.01009* (0.00)	−0.002103* (0.00)	0.013234* (0.00)	−0.000299 (0.70)	0.001852* (0.00)	−0.000120 (0.45)
Strategic indices	0.018772* (0.00)	0.019515* (0.00)	0.014239* (0.00)	0.014564* (0.00)	0.017204* (0.00)	−0.0036* (0.00)	0.017689* (0.00)	−0.00179* (0.01)	0.002322* (0.00)	−0.000079 (0.59)
Thematic indices	0.005665* (0.00)	0.004012* (0.00)	0.006028* (0.00)	0.004566* (0.00)	0.00889* (0.00)	−0.001236* (0.00)	0.008674* (0.00)	−0.000263 (0.69)	0.001346* (0.00)	−0.000060 (0.65)

Source: The authors.

Note:

The significantly positive p-values are marked with asterisk (*). However, the significantly negative values depicting the herd presence are marked bold and with asterisk (*).

All beta coefficients are positive, except in the case of intraday return in extreme down movement of the market, thereby, rejecting the herd behaviour in the stock market, which implies that investors act rationally and do not imitate the actions of other investors. Herd presence is rejected as the return dispersions tend to increase during extreme market movements. This indicates investors react to bad news much briskly rather than good news (Chen, 2013). The findings of the study are consistent with Yousaf et al. (2018), Chong et al. (2017), Kanojia et al. (2020b), Kumar et al. (2016) and Garg and Gulati (2013). May be due to the short time frame, investors herd and follow the trend during intraday trading. However, when they have sufficient time to ponder, deliberate and analyse, they might take relatively independent investment/trading calls based on their research and analytical acumen.

Moreover, the results of CSAD regression turn out to be somewhat different. Herd behaviour is witnessed in few instances in the Indian equity market. The findings of CSAD regression of Chang et al. (2000) conform the findings of Christie and Huang (1995) with an exception of daily return, absolute market return and squared market return, which are presented in Table 4.

Table 4.

CSAD Regression.

	Daily Return			Overnight Return			Intraday Return			Absolute Market Return			Squared Market Return
	R	AR	SR	R	AR	SR	R	AR	SR	R	AR	SR	R	AR	SR
Broad market indices	0.063* (0.00)	−0.048 (0.19)	3.802* (0.00)	0.030 (0.32)	0.002 (0.97)	4.450* (0.00)	0	0.271* (0.00)	−1.372* (0.00)	0	−0.046 (0.22)	3.982* (0.00)	0	−0.491 (0.78)	34.053 (0.46)
Sectoral indices	0.057* (0.00)	0.167* (0.00)	2.894* (0.00)	0.059* (0.00)	0.137* (0.00)	3.666* (0.00)	0	0.287* (0.00)	−0.481* (0.00)	0	0.161* (0.00)	3.134* (0.00)	0	0.452* (0.00)	12.166* (0.00)
Strategic indices	−0.003 (0.65)	0.622* (0.00)	1.516* (0.00)	0.020* (0.00)	0.640* (0.00)	1.494* (0.00)	0	0.309* (0.00)	0.127 (0.21)	0	0.371* (0.00)	0.662* (0.00)	0	0.884* (0.00)	−3.823* (0.00)
Thematic indices	0.054* (0.03)	−0.031* (0.00)	3.881* (0.00)	0.053* (0.00)	0.046* (0.00)	4.160* (0.00)	0	0.168* (0.00)	0.340* (0.00)	0	−0.018* (0.07)	3.996* (0.00)	0	0.168* (0.00)	18.966* (0.00)

Source: The authors.

Note:

• R stands for return

• AR stands for absolute return

• SR stands for squared return

• The significantly positive p-values are marked with asterisk (*). However, the significantly negative p-values depicting the herd presence are marked bold and with asterisk (*).

Herding is witnessed with significantly negative beta coefficients. However, herd presence is not found when beta coefficients are significantly positive. It is observed that herd presence is not validated and witnessed in the Indian equity market as all the coefficients, except few, are significantly positive. Although, a few coefficients of intraday return and squared market return are significantly negative. Both models observed consistent results thereby rejecting any herd presence in the Indian stock market during the selected sample period.

Conclusion

With the passage of time, the Indian financial sector has achieved great heights of growth, due to which, it has become an attraction for many international investors. Following the crowd and ignoring one’s own rationality constitutes herding. Herding exists because investors prefer to act in accordance with the majority rather than to stand alone. The absence of herding implies that investors perform rationally. However, this study aims to investigate the irrationality of investors by examining herd behaviour in the Indian equity market using broad market indices, sectoral indices, strategic indices and thematic indices of NSE. The calculated results failed to provide any evidence in favour of herding and support the theory of efficient market hypothesis. The results indicate that the Indian investors do not support herd mentality, rather take investment decisions on their own. Herding is an attribute of extreme market declines only.

The study has several implications for the policymakers, investors and regulators of the market as it would be helpful in correct valuation of stocks. These results can enable policy makers to design the reforms in the market microstructure.

Presence of herding in intraday return would help the arbitragers to act upon and book profits during short run caused due to price disequilibrium. In contrary to this, such opportunity is not available in daily, overnight, absolute and squared market returns. The reasons for absence of herding in the Indian stock market can be due to the participation of active traders in the market, who are institutional investors. These investors/traders drive the market as they may have the access to private information well before it reaches the retail investors. Such informational asymmetry may be the reason behind the absence of herding in the Indian equity market. Regulators are taking out various measures like improving transparency, full disclosure with respect to trading patterns, increase in trading hours, etc., which shall help to improve information symmetry in the market, resulting in informed decision making by the investors.

From the future scope point of view, this study examines the herd behaviour only in equity stocks, however, herd behaviour can also be examined for commodities, currencies, cryptocurrencies and derivatives.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

References

Abd-Alla

M. H.

(2020). Sentimental herding: The role of COVID-19 crisis in the Egyptian stock market. Copernican Journal of Finance & Accounting, 9(3), 9–23. 10.12775/CJFA.2020.009

Banerjee

A. V.

(1992). A simple model of herd behavior. The Quarterly Journal of Economics, 107(3), 797–817. 10.2307/2118364

Bikhchandani

, Hirshleifer

, & Welch

(1992). A theory of fads, fashion, custom, and cultural change as informational cascades. Journal of Political Economy, 100(5), 992–1026. 10.1086/261849

Bikhchandani

, Hirshleifer

, & Welch

(1998). Learning from the behavior of others: Conformity, fads, and informational cascades. Journal of Economic Perspectives, 12(3), 151–170. https://dx-doi-org-s.web.bisu.edu.cn/10.1257/jep.12.3.151

Bikhchandani

, & Sharma

(2000). Herd behavior in financial markets. IMF Staff Papers, 47(3), 279–310. https://www.imf.org/external/pubs/ft/wp/2000/wp0048.pdf

Calvo

G. A.

, & Mendoza

E. G.

(2000). Rational contagion and the globalization of securities markets. Journal of International Economics, 51(1), 79–113. 10.1016/S0022-1996(99)00038-0

Chang

E. C.

, Cheng

J. W.

, & Khorana

(2000). An examination of herd behavior in equity markets: An international perspective. Journal of Banking & Finance, 24(10), 1651–1679. 10.1016/S0378-4266(99)00096-5

Chen

(2013). Do investors herd in global stock markets? Journal of Behavioral Finance, 14(3), 230–239. 10.1080/15427560.2013.819804

Chevalier

, & Ellison

(1999). Are some mutual fund managers better than others? Cross-sectional patterns in behavior and performance. Journal of Finance, 54(3), 875–899. 10.1111/0022-1082.00130

10.

Choi

K. H.

, & Yoon

S. M.

(2020). Investor sentiment and herding behavior in the Korean stock market. International Journal of Financial Studies, 8(2), 34. 10.3390/ijfs8020034

11.

Chong

T. T. L.

, Liu

, & Zhu

(2017). What explains herd behavior in the Chinese stock market? Journal of Behavioral Finance, 18(4), 448–456. 10.1080/15427560.2017.1365365

12.

Christie

W. G.

, & Huang

R. D.

(1995). Following the Pied Piper: Do individual returns herd around the market? Financial Analysts Journal, 51(4), 31–37. 10.2469/faj.v51.n4.1918

13.

Dhall

, & Singh

(2020). The COVID-19 pandemic and herding behaviour: Evidence from India’s stock market. Millennial Asia, 11(3), 366–390. 10.1177/0976399620964635

14.

Economou

(2019). Herding in frontier markets: Evidence from the Balkan region. Review of Behavioral Finance, 12(2), 119–135. 10.1108/RBF-08-2018-0090

15.

Fama

E. F.

(1970). Efficient capital markets: A review of theory and empirical work. The Journal of Finance, 25(2), 383–417. 10.2307/2325486

16.

Garg

, & Gulati

(2013). Do investors herd in Indian market. Decision, 40(3), 181–196. 10.1007/s40622-013-0015-z

17.

Goldbaum

(2008). Coordinated investing with feedback and learning. Journal of Economic Behavior & Organization, 65(2), 202–223. 10.1016/j.jebo.2004.10.013

18.

Hong

, Kubik

, & Solomon

(2000). Security analysts’ career concerns and herding of earnings forecasts. RAND Journal of Economics, 31, 121–144. https://www-jstor-org-443.web.bisu.edu.cn/stable/2601032

19.

Kahneman

, & Tversky

(1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the Econometric Society, 47, 263–291. 10.1142/9789814417358_0006

20.

Kanojia

, Singh

, & Goswami

(2020a). Behavioural biases and trading volume: Empirical evidence from the Indian stock market. In Shrivastava

A. K.

, Sudhir

, Mohapatra

A. K.

, & Ram

(Eds.), Advances in management research innovation and technology (pp. 115–126). Taylor and Francis Publications.

21.

Kanojia

, Singh

, & Goswami

(2020b). Impact of herding on the returns in the Indian stock market: An empirical study. Review of Behavioral Finance, 14(1), 115–129.

22.

Katarachia

, Pitoska

, Giannarakis

, & Poutoglidou

(2018). The drivers of corporate governance disclosure: The case of Nifty 500 index. International Journal of Law and Management, 60(2), 681–700. 10.1108/IJLMA-02-2017-0020

23.

Kumar

, Bharti

, & Bansal

(2016). An examination of herding behavior in an emerging economy–A study of Indian stock market. Global Journal of Management and Business Research, 16(5), 57–64. https://dx-doi-org-s.web.bisu.edu.cn/10.13140/RG.2.2.22754.58562

24.

Luu

Q. T.

, & Luong

H. T. T.

(2020). Herding behavior in emerging and frontier stock markets during pandemic influenza panics. The Journal of Asian Finance, Economics and Business, 7(9), 147–158. 10.13106/jafeb.2020.vol7.no9.147

25.

Mertzanis

, & Allam

(2018). Political instability and herding behaviour: Evidence from Egypt’s stock market. Journal of Emerging Market Finance, 17(1), 29–59. 10.1177/0972652717748087

26.

Nofsinger

J. R.

, & Sias

R. W.

(1999). Herding and feedback trading by institutional and individual investors. The Journal of Finance, 54(6), 2263–2295. 10.1111/0022-1082.00188

27.

Poshakwale

, & Mandal

(2014). Investor behaviour and herding: Evidence from the National stock exchange in India. Journal of Emerging Market Finance, 13(2), 197–216. https://dx-doi-org-s.web.bisu.edu.cn/10.1177/0972652714541341

28.

Rashid

, Fayyaz

, & Karim

(2019). Investor sentiment, momentum, and stock returns: An examination for direct and indirect effects. Economic Research [Ekonomska istraživanja], 32(1), 2638–2656. 10.1080/1331677X.2019.1650652

29.

Scharfstein

D. S.

, & Stein

J. C.

(1990). Herd behavior and investment. The American Economic Review, 80(3), 465–479.

30.

Shiller

R. J.

(2003). From efficient markets theory to behavioral finance. Journal of Economic Perspectives, 17(1), 83–104. https://dx-doi-org-s.web.bisu.edu.cn/10.1257/089533003321164967

31.

Swank

, & Visser

(2008). The consequences of endogenizing information for the performance of a sequential decision procedure. Journal of Economic Behavior & Organization, 65(3–4), 667–681. 10.1016/j.jebo.2005.11.003

32.

X. V.

, & Phan

D. B. A.

(2019). Herd behavior and idiosyncratic volatility in a frontier market. Pacific-Basin Finance Journal, 53, 321–330. 10.1016/j.pacfin.2018.10.005

33.

West

(1988). Bubbles, fads and stock price volatility tests: A partial evaluation. Journal of Finance, 43(3), 639–656. 10.1111/j.1540-6261.1988.tb04596.x

34.

Yousaf

, Ali

, & Shah

S. Z. A.

(2018). Herding behavior in Ramadan and financial crises: The case of the Pakistani stock market. Financial Innovation, 4(1), 1–14. 10.1186/s40854-018-0098-9