INDIAN JOURNAL OF FINANCE AND BANKING 9(1) (2022), 164-176 164 FINANCE AND BANKING IJFB VOL 9 NO 1 (2022) P-ISSN 2574-6081 E-ISSN 2574-609X Available online at https://www.cribfb.com Journal homepage: https://www.cribfb.com/journal/index.php/ijfb Published by CRIBFB, USA THE ALTERED VOLATILITY SPILLOVER SEQUENCE UNDER COVID-19: INDIAN SECTORAL INDICES IMPACT DEIBOLD YILMAZ INDEX Kunwar Sanjay Tomar (a)1 (a) Professor, School of Management, FOSTIIMA, New Delhi, India; E-mail: sanjay.tomar@fostiima.org A R T I C L E I N F O Article History: Received: 01 January 2022 Accepted: 28 February 2022 Online Publication: 19 March 2022 Keywords: Diebold and Yilmaz Index Volatility Spillover, Bombay Stock Exchange Sector Indices, COVID- 19 JEL Classification Codes: G01, G11, D81, D85 A B S T R A C T The Industrial sectors have their unique place in the economic interlinkage. The sectoral valuation reflected by each sector indices shows how each sector responds to different events. The exogenous event Covid-19 impact has been differential due to the impact of lockdown and other Covid-19 appropriate restrictive measures. The present paper examines the change in the volatility spillover induced by Covid- 19. The study uses daily sectoral indices data from India's oldest exchange, the Bombay Stock Exchange. Data from January 2010 to November 2020 has been split into four subgroups to find how COVID-19 has affected the volatility spillover using the Diebold and Yilmaz Index. Ranks have been assigned to find the change in the four periods' volatility to the volatility spillover's magnitude and direction. The impact of the COVID-19 is strong enough to change the volatility spillover, which followed a system. Capital Goods volatility increased three times. At the same time, the Auto sector becomes a volatility receiver instead of the net volatility dispenser, from 2.5% before COVID-19 to -3.39% after COVID-19 lockdown. Bankex remains unaffected by Covid-19. © 2022 by the authors. Licensee CRIBFB, USA. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). INTRODUCTION The COVID-19 seems to have impacted sectors like hospitality, manufacturing, and services industry almost immediately, followed by other sectors. Hence volatility transmission also should have followed the same sequence. A case in point is the Information technology sector. The shifting of “work from home” would have made no significant difference to this sector's volatility. However, our findings show a complete upsurge of the volatility sequencing. COVID-19: The Contextual Background COVID-19 is a rare event. It needs intense scrutiny, rightly termed a genuine exogenous shock (Ramelli & Wagner, 2020). Mainly so as the COVID-19 with 2.5 million deaths (March 1 2021) is next only to Spanish Flu in terms of fatality. The other pandemics and epidemics like 1957-58 H2N2, H3N2 1968, 2009-10 Swine Flu,2012 MERS, 2014-16 Ebola did not disrupt the globe as COVID-19 has. One common strand most researchers affirm today is how gargantuan COVID-19 is. Compared to COVID, the Spanish Flu killed nearly 50 million (Zimmer & Burke, 2009) worldwide, while around 20 million in India(Chandra & Kassens-Noor, 2014). The research on Spanish Flu is not new, neither for the finance field nor for medical sciences. Medical research has been active in the last decade, calling the Spanish flu virus the “mother of all pandemics”(Taubenberger & Morens, 2006). Much of such research extensively studied the COVID-19 type pandemic through the study of the Spanish Flu (Boëlle et al., 2011; Martini et al., 2019). Baker et al. (2020) compared the financial markets under the present pandemic and other such epidemic effects. They confirm through their study that COVID-19 pandemic is most severe in its impact in the entire time frame from 1900 onwards. The impact of COVID-19 is almost double that of the Spanish Flu, as per Baker. More work hence is required to 1Corresponding author: ORCID ID: 0000-0003-0166-3749 © 2022 by the authors. Hosting by CRIBFB. Peer review under responsibility of CRIBFB, USA. https://doi.org/10.46281/ijfb.v9i1.1660 To cite this article: Tomar, K. S. (2022). THE ALTERED VOLATILITY SPILLOVER SEQUENCE UNDER COVID-19: INDIAN SECTORAL INDICES IMPACT DEIBOLD YILMAZ INDEX. Indian Journal of Finance and Banking, 9(1), 164-176. https://doi.org/10.46281/ijfb.v9i1.1660 http://creativecommons.org/licenses/by/4.0/) http://creativecommons.org/licenses/by/4.0/) https://doi.org/10.46281/ijfb.v9i1.1660 https://orcid.org/0000-0003-0166-3749 Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 165 understand many aspects of such events. The essential objectives are interlinkage of volatility, global penetration, and the ripple effect to study the financial meltdowns. As we move through the COVID-19 much research is being conducted to understand the financial markets' response to such black swan (Loginov & Heywood, 2020) events. What seems to be even more challenging is the economic recovery mechanism. Research studies on the impact of a pandemic, business cycles, financial crisis focus on the genesis and influence of channelization. However, the system for research to study volatility inflicted by COVID-19 creates its peculiar mechanism. On the one hand, we grapple with the battle to nurture back the economy to normal, while on the other hand, the COVID-19 does not seem to end. The present pandemic of COVID-19 is most intense in its impact. The spread of pandemic upsurge covering all the countries is another fact that the global community is coming to terms. The question now is whether this phase of uncertainty will end or whether COIVD 19 has become a more regular part of our lives. For the researchers, financial market and economic regulators, business managers, and fund managers, this phase of COVID- 19, the battle of various vaccines and mutating viruses, is the test of time. Researches need to find insights into this COVID- 19 pandemic. Why Sector Level Study of Volatility? The present study adds to the literature by opening the dimension of sector behaviour peculiar to COVID-19. The equity sector indices reflect the expected future earnings and valuation. The valuation differs across the sectors. The importance of the sectors as economic focal point has been changing dynamically as India moves from underdeveloped to developed economy. Over the years, the Indian economic structure has changed. Labour-intensive manufacturing has changed to a more service-based economy, changing the real wage to the rental price of capital ratio. The labour intensity has changed from 1.45 in the 1980s to 0.33 in the 2000s. (Economics & Series, 2014). The changing structure of the Indian economy itself needs regular assessment to assess the sector level data. With the COVID-19, a new window has opened to find how the sectors have behaved—especially the labour-intensive economic activities. COVID-19 has led to significant disruption for the factory workers. Nearly 600 million workers migrated internally in India due to COVID-19 (Covid-19 fallout: How the pandemic displaced millions of migrants - News Makers News - Issue Date: January 11, 2021, n.d.) Although the scenario before COVID-19 was not too good either. As per the MOC (Ministry of Commerce), 2019-20 had already seen a flat growth in the eight core industries (Coal, Crude Oil, Natural Gas, Refinery Products, Fertilizers, Steel, Cement, and Electricity)2. Sector level study need to be studied, with regards to how the lockdown has impacted certain sectors more than the others. How the sectors have transmitted the risk can be assessed. Interest in sector-level associations and functionality has been a popular topic for research (Bahmani-Oskooee & Saha, 2019) . Narrowing to a more specific study of volatility spillover and connectedness, Gabauer et al. (2020) research is worth mentioning. Gabauer et al. (2020) has undertaken a more recent study to find the volatility spillover connectedness for Indian sectors. The unique relation supply demand relation sectors have many not necessarily transmit volatility in the order expected. Gabauer et al. (2020) consider the leading/lagging connectedness the driver of many critical growth-oriented decisions. Their study shows that this connectedness varied in India's previous crisis, mainly 2008, Inflation of 2011, national elections, and Demonetization of 2016. They tried to answer how this change occurred. The policy changes in particular between 2014-19 with the focus on “Making in India” and several initiatives to increase employment, regulate banking (the recent past has seen many mergers of Public sector banks) reflected in their study. Like the Ghatziantonion study, this paper tries to understand the system's volatility spillover connectedness the COVID-19. We use the Diebold and Yilmaz (2012) Index (DYI) with a volatility measure of Parkinson (Parkinson, 1980). The DYI can see the volatility spillover over different asset class portfolios and between the sectors. Their Index bypasses the controversial issues associated definition and existence of episodes of “contagion” or “herd behaviour” as per Diebold and Yilmaz. The connectedness, by definition, is a linkage or relatedness of the components under study. The (Xiao & Huang, 2018) DYI is used by many authors. DYI has also led to more augmented methods like by Gabauer (2020); Antonakakis et al. (2018). Xiao and Huang compare and contrast the different methods used in measuring connectedness. Their study classifies the DYI based on different methods based on the volatility spillover and the system's contribution. They find the DYI 2012 appropriate when the whole system is the understudy for the volatility of the variable spillover connectedness and not merely the correlation. The direction of volatility spillover with pair-wise calculation makes the DYI intuitively superior to other methods. We can understand “which sector gives and receives the volatility spillover?”. More important feature of DYI 2012 methodology is to find the primary variable dispensing the highest volatility spillover and hence the variable which takes the central role in the system-wide volatility spillover connectedness. LITERATURE REVIEW Researchers agree that the present pandemic is comparable to the worst financial market meltdowns. The financial market downturn inflicted by COVID-19 is unparalleled in its magnitude, penetration, and severity. The Spanish Flu, which killed 50 million (The Spanish Flu (1918-20): The global impact of the largest influenza pandemic in history - Our World in Data, n.d.) People as compared to 2.5 million by COVID-19, had a much lesser impact on financial markets and economy. Baker and others find no other epidemic and pandemic having such intense effect on the economy and financial markets as COVID- 19. They write in their white paper that the Spanish Flu impact was modest compared to COVID-19. (Baker et al., 2020a). Perhaps that brings under scrutiny the Governmental policies to check the COVID-19. Questions being asked such as “Could the cost of battling COVID-19 have been much lesser?”.(Flatten the Coronavirus Curve at a Lower Cost - WSJ, 2 https://pib.gov.in/PressReleasePage.aspx?PRID=1601314 Ministry of Commerce and Industry India https://pib.gov.in/PressReleasePage.aspx?PRID=1601314 Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 166 n.d.)(Coibion et al., 2020). The lockdown, as a policy decision to check COVID-19, itself affected some sectors more than the others. Notably, the labour-intensive sectors were hit most by COVID-19 pandemic (Chaudhary et al., 2020). These measures containing the spread of Corona required immediate shut down of manufacturing and other such labour-intensive sectors. In India, the labour working in different manufacturing zones is spread across the country; labour to such sectors is predominantly served by the two most populous states of UP and Bihar. Most of these day workers had no other option but to migrate to their native place. The triggering ripple effects led to massive internal migration in India (more details given earlier in this article). These ripple effects should reflect in the financial market’s valuation mechanism. Hence, volatility would follow a sequence linked to the supply chain system. Banks, the Power sector, Logistics are some sectors that continued functioning. Thereby immune to the lockdown but certainly affected by the social distancing and even spread of infection. Such differential sector level operations should reflect in the financial market information processing. There is a need to find the impact of financial markets and their sequence. Such asymmetric connections in sectors become vital for fund managers, regulators, and business managers. The use of DYI is apt for such event as COVID-19, as discussed above. The method has been used by researchers in similar financial and economic crisis previously. Sehgal et al. (2015) using DYI find a change in the directional flow from the U.S. to Europe, varying as the stages of the 2008 crisis deepened. Accentuating the importance of connectedness measure even more, show how equity markets connectedness shows a robust geographical component, not found in the case for Bond markets. Their study reaffirms stronger international interlinkage in the “Great Financial crisis.” Such volatility spillover connectedness has led to a series of studies that try to find the linkage among global institutions and markets. Khan and others' complex network method finds structural change through "node changes, clustering, and homogeneity" in the world market (Aslam et al., 2020). It is not surprising that in COIVD 19, the major banks' connectivity increased, and so did the spillover density. The author puts it more effectively using the word "unprecedented interconnectivity" in COVID- 19 using DYI (Baumöhl et al., 2020). Researchers define COVID-19 as a pure “exogenous factor.” Unlike other crises such as political, economic, and financial, COVID-19 is an exact exogenous event to study the firm and industry interaction in such rare event. The infectious diseases earlier have been grossly underrated and slowly has crept into a more obscure event, as Ramelli and Wagner pointed out. Their paper quote “World Economic Forum’s Global Risk Report (2020)” (Hall, 2020) listed the infectious disease as the tenth item in order of impact strength. In their paper, Ramelli and Wagner cite how the “disaster literature” can explain the complexity of such events and how they relate to their future use (Ramelli & Wagner, 2020). The researchers have been quick to provide useful benchmarks for the steps taken by different governments to check COVID-19. An example of such a study is by Carletti et al. (2020). They find a three-month lockdown to reduce the profit by 10% of yearly GDP. The most vulnerable are the small and mid-size organisations (Carletti et al., 2020). The literature on COVID-19 can also be viewed as published initially at the beginning of the COVID-19. As WHO declared the pandemic, markets and governments acted. To this announcement by WHO, the researchers acted almost immediately. The researchers started publishing as early as March April 2020. Like Liu et al. (2020) published their work in April 2020. They effectively laid the composition for classifying the research of earlier similar studies on catastrophic events. Their study shows how more digitalised firms stood the test of the time to face COIVD 19 (Ding et al., 2020). Many such studies captured the data up to March 2020, publishing in July to September 2020. Some set of these studies compared the pandemic to other such crisis times. However, COVID-19 pandemic differed from other such events in many ways. Firstly, the unique aspect of the COVID-19 is its almost simultaneous global onset. The studies measuring the impact of COIVD on markets use many methods to find the specific features relating to the COVID-19 pandemic. The epidemics like Ebola, SARS MERS, have been concentrated more in certain geographical regions but not COIVD 19. The panic and fear impact of the pandemics have been studied relatively well (Long et al., 2021). Their study shows that the pandemic's impact is more on the emerging economies than on the developed markets. The studies also related to the regions and economic classification based on development. Albuquerque, Koskinen, Yang, and Zhang study and find that the companies high on the E.S. (environment and social) policies perform better than those with lower E.S. scores. They consider the COVID-19 pandemic to test the ESG (environment, social, and governance) theories (McWilliams & Siegel, 2001; Friedman, 1970) as opposed to the ESG) (Albuquerque et al., 2020). Liu and others study the COVID-19 impact on the 21 leading stock markets to show how significant this pandemic is. They use the event study with the cumulative abnormal return. The event study using especially CAR (Cumulative abnormal return) remains the most preferred method used by the researchers to study COIVD 19. Liu and others published in June 2020 the short-term impact of the COVID-19. They also study the sectoral indices to find the impact of the COVID- 19. Their study shows that pharmaceuticals, I.T. ware favored by the investors, while transport, lodging, and catering were negatively affected (Liu et al., 2020). The impact of COVID-19 is hence asymmetric. In the economy, the sectoral distribution of the COVID-19 pandemic also attracted studies. Ten sectors are studied by Liew and Puah (2020). Liew and Puah (2020) find the OCIVD 19 effect on the Shanghai stock exchange and sectoral indices. They find that I.T. and Telecom were more immune to the COVID-19 effect. Studies focusing on Indian markets and COIVD 19: The studies using the Indian NSE/BSE or primary markets found a substantial drop in the markets on January 20, 2020. H. Liu et al., (2020) based on abnormal and cumulative results returns. The volatility spillover based on market size shows considerable volatility from mid-cap to small-cap and primary Index of Bombay stock exchange. Trabelsi and others use Indian financial markets and Gold for portfolio optimisation during the COVID-19 times. Bora and Basistha study the COVID-19 effect on the Indian stock market using the GJH GARCH model. They find a significant effect of volatility on BSE. NSE was not affected with the same magnitude. This Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 167 study, which takes the data from September 3 to July 10, 20020, finds that the upward trend started in their sample period (Bora, Debakshi, & Basistha, 2020). Salisu et al. (2020) find that the emerging markets are affected more than the developed markets. They used 24 emerging markets and 21 (India as one of them) developed markets. Using out of sample and full sample data, they conclude that government policies have no effect on uncertainty from COVID-19. They used the Equity Market Volatility Infectious Disease Tracker (EVM). Yousaf and Ali trace the high-frequency information transmission among the cryptocurrencies using VAR-DCC- GARCH (Yousaf & Ali, 2020). Their finding shows the unidirectional spillover from Ethereum to Bitcoin and Bitcoin to Litecoin. The literature above has summarised the COVID-19 research, which sought to bring out different aspects of the COVID-19 on financial markets. The data used, methodology, statistical tools, graphs, and software. Put together, these studies point out the validity of finding more insights into such catastrophic events. Medical science and researchers have been pointing out the possibility of outbreak of events such as COVID-19. Had these forewarnings have been taken more seriously the COVID-19 could have been better tackled. On the ground level, the large internal migration of the labour in India and the loss of livelihood for the day worker is a massive hit to the “unfortunate bottom of the pyramid.” The digitalised India quickly responded by giving relief packages (May 15 2020)of USD 260 billion (India- Measures in response to COVID-19 - KPMG Global, n.d.), saving the worker. The unemployment rate of 23.52% in (• India: unemployment rate due to COVID-19 | Statista, n.d.) April 2020 (which now as of March 2021 is 6.53%) had been severe, making them walk hundreds of kilometers. The present work finds this gap in understanding the volatility spillover under COVID-19 at the sector level. DATA AND METHODOLOGY Model Used The use of the volatility spillover and connectedness approach by Diebold and Yilmaz can give an insight into the mechanism of volatility spillover. The past data of the financial market would show a system of connectedness based on historical data. The upheaval brought by COVID-19 in the economy and markets should follow the same system of connectedness. The out-of-sample data would affirm such a system. The data set before the COVID-19 can be seen as a benchmark to compare the CORONA period as out of sample data for comparison. Although many researchers have used connectedness and volatility spillover, no such study uses this methodology to compare the sub-periods post and previous to the COVID-19. The primary motivation for using the Diebold and Yilmaz Index and other derivations of their method by researchers is presented under. In their paper "Better to give than receive: Predictive directional measurement of volatility spillover," (Diebold & Yilmaz, 2012) extends their Index further. D.Y. spillover index is an output of variance decomposition with N-variable vector autoregression. Familiar terrain for researchers. The primary focus of the D.Y. is on the total spillover in a somewhat simplified VAR model. The Cholesky factor orthogonal drives the potential order-dependent results. As directional spillover is measured in a generalised VAR framework, it eliminates the dependency on ordering results. An N-variable VAR (p) (covariance stationery) 𝑥𝑡 = ∑ ∅𝑖 𝑝 𝑖=1 𝑥𝑡−𝑖 + 𝜀𝑖. The identically distributed disturbance vector is represented by 𝜀𝑖(0, ∑). The NxN coefficient matrices Ai obeys the recursion 𝐴𝑖 = ∅1𝐴𝑖−1 + ∅2𝐴𝑖−2 + ⋯ ∅𝑝𝐴𝑖−𝑝, with A0 an NxN identity matrix and 𝐴𝑖 = 0 for i<0. The dynamics of the system build on the moving average coefficients. An important part is the "system shocks." These system shocks segregate in various components based on variance decomposition. Such as variance decomposition impulse response. The variance decompositions make the fractions of H-step ahead error variance in forecasting Xj, Ɐj≠i, for each i. While the VAR innovations are contemporaneously correlated, calculating variance decompositions requires orthogonal innovations. Cholesky factorisation achieves orthogonality as the identification method. Variance decomposition depends on the variable orders act. DIY solves this by using VAR generalised framework. Here, the generalised approach allows correlated shocks and explains the past observed error distribution. This is done instead of orthogonalising the shocks. This way, the total contribution to the variance of forecast error (Row sum of the variance decomposition table) may not be equal to unity. Such exploratory power allows correlated shocks. The fraction of the H-step ahead error variance in forecasting is forecasting, is "own variance shares" for xi due to shocks to xi, for i=1,2,…, and cross variance shares, or spillovers, to be the fractions of the H-step ahead error variances in forecasting xi, due to shocks to xj, for i,j=1,2…, N, such that i≠j. The expression denoting KPPS H-step ahead forecast error variance decomposition 𝜃𝑖𝑗 𝑔(𝐻), for H = 1,2…, we have 𝜃𝑖𝑗 𝑔(𝐻) = 𝜎𝑖𝑖 −1 ∑ (�́�𝑖𝐴ℎ∑𝑒𝑗)^2𝐻−1 ℎ=0 ∑ (�́�𝑖𝐴ℎ∑𝑒𝑖)𝐻−1 ℎ=0 (1) Here the Variance matrix for error vector ε, is ∑. The standard deviation of the error term for the ith equation is σij. Whereas ej is the selection vector with one as the ith element and zero otherwise. As expressed earlier ∑ 𝜃𝑖𝑗 𝑔𝑁 𝑗=1 (𝐻) ≠ 1, variance decomposition table each element in a row is not equal to one. For calculating the spillover index, and utilising the information through variance decomposition matrix, normalising each entry of the variance decomposition matrix by the sum, can be expressed as Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 168 �̃�𝑖𝑗 𝑔 (𝐻) = 𝜃𝑖𝑗 𝑔 (𝐻) ∑ 𝜃 𝑖𝑗 𝑔𝑁 𝑗=1 (𝐻) (2) By construct ∑ �̃�𝑖𝑗 𝑔𝑁 𝑗=1 (𝐻) = 1 and ∑ �̃�𝑖𝑗 𝑔𝑁 𝑗=1 (𝐻) = 𝑁. The total spillover volatility index is constructed by variance decomposition volatility contribution from the KPPS. �̃�𝑖𝑗 𝑔 = ∑ �̃�𝑖𝑗 𝑔 (𝐻)𝑁 𝑖,𝑗=1,𝑖≠1 ∑ �̃� 𝑖𝑗 𝑔 (𝐻)𝑁 𝑖,𝑗=1 x 100 = ∑ �̃�𝑖𝑗 𝑔 (𝐻)𝑁 𝑖,𝑗=1,𝑖≠1 ∑ �̃� 𝑖𝑗 𝑔 (𝐻)𝑁 𝑖,𝑗=1 x 100 (3) The spillover of volatility shocks across variables contributes to the total forecast error variance estimated by the total spillover index. Spillover of volatility contributed by variables in analysis to the total forecast error variance is measured by the total spillover index. More meaningful information is provided by the direction of the spillovers across variables. The generalised VAR approach enables us to provide this informative part. Much information can be extracted from the total volatility spillover index, and directional spillovers complete the total picture of volatility movement. The directional volatility spillovers through normalised elements of the general variance decomposition matrix hey can be expressed as 𝑆𝑖 𝑔(𝐻) = ∑ �̃�𝑖𝑗 𝑔 (𝐻)𝑁 𝑗=1,𝑗≠1 ∑ �̃� 𝑖𝑗 𝑔 (𝐻)𝑁 𝑗=1 x 100 (4) Measures B.I. directional volatility spillovers by variable i from all other variables j. This is generalised impulse responses, and variance decompositions are invariant to the ordering of variables 𝑆𝑖 𝑔(𝐻) = ∑ �̃�𝑖𝑗 𝑔 (𝐻)𝑁 𝑗=1,𝑗≠1 ∑ �̃� 𝑖𝑗 𝑔 (𝐻)𝑁 𝑗=1 (5) Net spillovers from variable i to all other variables j are expressed as 𝑆𝑖 𝑔(𝐻) = 𝑆𝑖 𝑔(𝐻) − 𝑆𝑗 𝑔(𝐻) (6) To find the net volatility spillover, we can calculate the difference between gross volatility shocks received and gross volatility shock transmitted from all other variables. The pair-wise variable i to j for volatility spillover is the difference between gross volatility shocks transmitted from variable i to j and that transmitted from j to i. Net pair-wise spillover, in addition to the net volatility spillover, make the interpretation much more effortless. Expressed as: 𝑆𝑖𝑗 𝑔(𝐻) = �̃�𝑖𝑗 𝑔 (𝐻) ∑ �̃� 𝑖𝑘 𝑔 (𝐻)𝑁 𝑘=1 − �̃�𝑗𝑖 𝑔 (𝐻) ∑ �̃� 𝑗𝑘 𝑔 (𝐻)𝑁 𝑘=1 x 100 (7) Data The data used is daily for the ten sectors and primary Index "Sensex." All the indices are from India's oldest and most popular Index, the "Bombay Stock Exchange." The time taken is from January 4 2010, up to November 2020. (The limitation of the time period has been the data availability, which for some of the Index starts from the date given). Like Diebold and Yilmaz for volatility Parkinson method has been used. Parkinson, (1980) volatility measure requires low and high Index values during the day for each Index. For some of the indexes, this data was not available (daily “High” and “Low”) for the period earlier to January 2010. Hence, such indexes were left out of the calculation. Analysis Methodology The analysis seeks to find and explain the volatility spillover peculiar to COVID-19. For this reason, Diebold and Yilmaz Index (referred hereafter as DYI) is used. The following sectors from BSE (Bombay Stock Exchange) are included; Automobile (Auto), Banks (Bankex), BSE (Sensex, the primary Index of BSE), Capital Good (CAP), Consumer goods (CD), Metal (Metal), Oil and Gas, Power, Reality, Technology (TECH). The (DYI) Diebold Yilmaz Index output provides total volatility spillover within the model and each sector associated with other sectors. The advantage of DYI is that it shows which sector is more or less volatile than other sectors with a directional flow of volatility. This is shown in each DYI by the row named as CTO and column as FROM. The column and row “From” “CTO” (contribution to others) give the volatility spillover received by each sector from the other sectors and disseminated to other sectors, respectively. The net volatility spillover row shows if the sector is the net receiver or provider of volatility spillover. Diagonal in each DYI model gives each sector’s volatility spillover. Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 169 The data from January 4 2010, to November 14 2020, is split into four parts. (i) Daily volatility from January 4, 2010, to November 14 2020, referred as S1 (ii) Daily volatility from January 4, 2010, to January 31 2019, referred as S2 (iii) Daily volatility from February 4 2019, to December 31 2019 (221 days) as Pre COVID-19 (iv) Daily volatility from January 1, 2020, to November 14 2020 (221 days) as during COVID-19 The above period group S1 constitutes the entire sample. The other three samples become the “In sample” subsets. The three-period groups, S2, Pre and During COVID-19, becomes the separate “Out of sample” sets. The time set S2, Pre COVID-19, and COVID-19 should have approximately the same volatility profile. The DYI are assigned ranks. The reasons are as follows:- DYI is calculated for each of the four periods. As these volatilities are in percentage, each sector role in volatility is known, making a comparison across four periods easy. For example, the Auto sector column “From” shows the volatility from other sectors. For the total period under study (which includes the COVID-19 period), volatility received by the Auto sector is 68.32%. This increases to 82.29% during the COVID-19. However, this shows the quantum jump in volatility due to COVID-19. It does not show if the Auto sector has become more or less volatile in the crisis than other sectors. The 13.97% increase in the Auto sector can be similar for all the sectors if all sectors receive the same exogenous impact. If we rank each sector for the volatility being received and transmitted, we can also know if a particular sector has become more or less volatile relative to other sectors. For this reason, the assigned ranks are summarised in Table 10. The rank of the Auto sector remains the fifth largest volatility receiver in all the three periods baring 221 days period before COVID-19, whereas it was the seventh-largest volatile receiver. It thereby becomes clear that Auto volatility reduced before the COVID-19 sub-sample of the period. Hence the percentage change in the volatility needs to be seen in comparison to the entire batch of sectors in comparison. Another dimension added by ranking is the relative increase or decrease of volatility as “receiver” or “transmitter”. The Auto sector volatility ranks as the receiver is maintained at 5th rank even in COVID-19 time, but the volatility transmission becomes seventh from sixth rank pre COVID-19. The transmission of volatility is reduced. This position is also seen in the column “Net”, which shows Auto sector transmit -3.39% volatility but reduces the overall rank to the seventh-highest transmitter. Plan of Analysis The descriptive statistics for the entire sample data of S1 is based on returns. Daily variance is used using the high and low prices of each sector used by Diebold and Yilmaz. The Parkinson (Parkinson, 1980) method is used for the DYI. For sector i on the day t we have 𝜎𝑖𝑡 2̃ = 0.361[𝑙𝑛(𝑃𝑖𝑡 𝑚𝑎𝑥) − 𝑙𝑛(𝑃𝑖𝑡 𝑚𝑖𝑛)]2 where 𝑃𝑖𝑡 𝑚𝑎𝑥 is the high and 𝑃𝑖𝑡 𝑚𝑖𝑛 is the low in the market i on day t Table 1. Discriptive Statistics for all the ten sectors and primary index Table 1 Auto Bank CAP CD Metal OG Power PSU Reality BSE Tech Mean 0.04% 0.06% 0.02% 0.08% -0.01% 0.02% -0.01% -0.02% -0.01% 0.04% 0.05% Standard Error 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% Median 0.10% 0.10% 0.00% 0.10% 0.00% 0.00% 0.00% 0.00% 0.10% 0.10% 0.10% Standard Deviation 1.40% 1.60% 1.50% 1.40% 1.70% 1.40% 1.30% 1.30% 2.00% 1.10% 1.20% Sample Variance 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% Kurtosis 821.30% 894.30% 656.30% 546.20% 289.60% 857.50% 439.10% 654.00% 204.10% 1395.60% 761.70% Skewness -16.10% -34.20% -25.00% -29.50% -12.60% -56.00% -35.70% -34.60% -29.70% -75.90% -44.30% Range 23.60% 27.50% 23.70% 20.30% 20.70% 21.70% 18.60% 19.90% 19.70% 22.10% 17.90% Date for Min Return 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 23-Mar-20 Minimum -13.40% -16.80% -14.90% -11.70% -11.90% -12.70% -8.40% -10.90% -10.90% -13.20% -9.60% Maximum 10.30% 10.70% 8.80% 8.60% 8.80% 9.10% 10.20% 9.00% 8.80% 9.00% 8.40% Sum 1.1629 1.5093 0.4137 2.1737 -0.2144 0.4866 -0.3138 -0.4061 -0.1426 1.0731 1.3097 Count 2698 2698 2698 2698 2698 2698 2698 2698 2698 2698 2698 The descriptive statistics sectors are based on returns for the total period S1 (table 1). This shows that the worst day of Indian financial markets occurred on March 23 2020. The highest fall is recorded for the Bank & Capital Goods sector with -16.8% and -14.9%. The least single-day fall was in the Power sector. The highest mean returns for the whole sample period are given by Consumer durable on a daily basis. The highest and the lowest standard deviance is shown by reality and Metal with 2% and 1.7%. The skewness, a measure of asymmetry threshold, is negative for all the sectors. Hence, the sectors' return is longer to the left side of the distribution than to the right. As the universal indices for markets in general, the high kurtosis is heavy-tailed.(“Coefficient of Skewness,” 2008) ANALYSIS Total Volatility Spillover Figure 1 compares the total sample size with the other three sets of the sample for the "Total volatility spillover." The sample consisting of the entire sample which also includes (blue colour) the COVID-19 volatility spillover. The volatility spillover reaches 82.5% during the COVID-19. Earlier to the COVID-19 period, the sample period of 2010-18 (figure 1) with the highest touching 78% compared to COVID-19 at 82.5%. Most of the earlier periods in the two graphs show how the total Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 170 system volatility spillover ranges between 67% to 77%. The upsurge of the volatility spillover hence becomes very clear. For the sample period of 221 days pre COVID-19, Figure 4, the volatility spillover’s highest point is 68.9% (approx.), while after Covid Figure 5 shows 81.8% volatility 55 60 65 70 75 80 85 10 12 14 16 18 20 Full Sample Non Coivd Total Volatility Spillover 2010-2020 Figure 1. Total Volatility spillover 2010-2020 Figure 2. Dynamic Total Connectedness Full Sample S1 Figure 3. 2010-18 Dynamic Total Connectedness S2 Figure 4. Pre COVID-19 Dynamic Total Connectedness Figure 5. COVID-19 Dynamic Total Connectedness The Total Connectedness The total volatility spillover, the Composite Index of various directional volatility spillovers, appears in the lowermost right corner of each index table. If one looks at Pre and during COVID-19 and the full sample (S1) index table, we can note that Total volatility spillover is 67.7%, 81.24%and 72.6%.This shows that the Covid-19 pandemic turbulence adds a 27% increase in volatility. Table 2. Data S1 Table 2 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech Mean -9.476 -9.355 -9.296 -9.087 -8.902 -9.373 -9.48 -9.599 -8.565 -10.073 -9.742 Variance 0.936 1.105 0.934 0.84 0.808 0.818 0.9 0.949 0.893 0.991 0.823 Skewness 0.365*** 0.392*** 0.380*** 0.447*** 0.253*** 0.471*** 0.332*** 0.372*** 0.271*** 0.487*** 0.450*** 0 0 0 0 0 0 0 0 0 0 0 Kurtosis 0.443*** 0.347*** 0.333*** 0.425*** 0.233** 1.081*** 0.543*** 0.439*** 0.202** 0.771*** 0.777*** 0 -0.001 -0.002 0 -0.022 0 0 0 -0.042 0 0 JB 82.128*** 82.816*** 77.315*** 110.128** 34.935*** 231.362** *82.824*** 84.061*** 37.557*** 173.719** 158.944*** 0 0 0 0 0 0 0 0 0 0 0 ERS -8.022*** -2.952*** -2.335** -5.475*** -8.616*** -5.953*** -2.473** -4.842*** -3.212*** -7.030*** -3.492*** 0 -0.003 -0.02 0 0 0 -0.013 0 -0.001 0 0 Q(20) 2807.845* 4104.888* *2086.891* *1787.858* 1885.059* *1826.328* *2103.494* 2104.510* 1133.040* *3768.076* 1707.907** Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 171 0 0 0 0 0 0 0 0 0 0 0 Q2(20) 2190.766* 3312.629* *1721.525* *1589.049* 1354.105* *1274.887* *1770.456* 1518.099* 888.473** *3023.454* 1274.023** 0 0 0 0 0 0 0 0 0 0 0 LM(20) 306.420** 425.344** *230.745** *238.739** 254.348** *236.133** *270.231** 273.705** 146.375** *431.797** 202.161*** 0 0 0 0 0 0 0 0 0 0 0 Auto 1 0.651 0.588 0.476 0.584 0.55 0.578 0.599 0.533 0.725 0.478 Bankex 0.651 1 0.638 0.484 0.549 0.54 0.59 0.659 0.546 0.816 0.486 CAP 0.588 0.638 1 0.477 0.533 0.536 0.662 0.63 0.561 0.649 0.416 CD 0.476 0.484 0.477 1 0.425 0.434 0.453 0.464 0.488 0.508 0.369 Metal 0.584 0.549 0.533 0.425 1 0.555 0.612 0.674 0.526 0.591 0.443 OilGas 0.55 0.54 0.536 0.434 0.555 1 0.583 0.719 0.507 0.639 0.469 Power 0.578 0.59 0.662 0.453 0.612 0.583 1 0.746 0.57 0.604 0.435 PSU 0.599 0.659 0.63 0.464 0.674 0.719 0.746 1 0.579 0.643 0.445 Reality 0.533 0.546 0.561 0.488 0.526 0.507 0.57 0.579 1 0.552 0.393 BSE 0.725 0.816 0.649 0.508 0.591 0.639 0.604 0.643 0.552 1 0.607 Tech 0.478 0.486 0.416 0.369 0.443 0.469 0.435 0.445 0.393 0.607 1 Table 3. Data S2 Table 3 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech Mean -9.571 -9.429 -9.31 -9.065 -8.983 -9.423 -9.542 -9.689 -8.568 -10.127 -9.801 Variance 0.836 0.978 0.901 0.789 0.757 0.756 0.871 0.875 0.877 0.898 0.748 Skewness 0.169*** 0.211*** 0.237*** 0.374*** 0.184*** 0.294*** 0.266*** 0.282*** 0.203*** 0.227*** 0.220*** Kurtosis -0.132 -0.205** -0.058 0.165 0.005 0.677*** 0.268** 0.09 0.112 -0.149 -0.047 -0.192 -0.031 -0.617 -0.118 -0.901 0 -0.017 -0.361 -0.267 -0.136 -0.694 JB 12.418*** 20.774*** 21.498*** 55.265*** 12.807*** 75.627*** 33.353*** 30.648*** 16.658*** 21.525*** 18.438*** ERS -7.896*** -2.853*** -2.100** -5.160*** -8.338*** -5.333*** -2.352** -4.947*** -2.966*** -6.579*** -3.240*** Q(20) 1398.319* 2235.424* 1493.624** 1058.597** 1029.662* 1011.197** 1366.441** 1140.460* 726.933*** 2347.848** 780.619*** Q2(20) 1283.337* 2077.128* *1298.618* *1005.396* 780.192** *794.434** *1214.136* 882.724** *610.523** *2173.183* 728.878*** LM(20) 191.234** 311.336** *185.397** *166.934** 160.068** *162.585** *193.686** 184.902** *98.927*** 335.616** 117.595*** Auto 1 0.627 0.577 0.463 0.541 0.5 0.563 0.558 0.534 0.711 0.439 Bankex 0.627 1 0.639 0.463 0.52 0.496 0.584 0.639 0.533 0.794 0.419 CAP 0.577 0.639 1 0.435 0.516 0.497 0.67 0.614 0.548 0.624 0.381 CD 0.463 0.463 0.435 1 0.419 0.396 0.445 0.463 0.466 0.47 0.331 Metal 0.541 0.52 0.516 0.419 1 0.518 0.599 0.648 0.532 0.558 0.406 OilGas 0.5 0.496 0.497 0.396 0.518 1 0.555 0.684 0.481 0.593 0.426 Power 0.563 0.584 0.67 0.445 0.599 0.555 1 0.735 0.577 0.588 0.404 PSU 0.558 0.639 0.614 0.463 0.648 0.684 0.735 1 0.581 0.604 0.397 Reality 0.534 0.533 0.548 0.466 0.532 0.481 0.577 0.581 1 0.528 0.366 BSE 0.711 0.794 0.624 0.47 0.558 0.593 0.588 0.604 0.528 1 0.567 Tech 0.439 0.419 0.381 0.331 0.406 0.426 0.404 0.397 0.366 0.567 1 Table 4. Data Pre - COVID-19 Table 4 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech Mean -9.261 -9.575 -9.426 -9.514 -8.673 -9.358 -9.49 -9.391 -8.843 -10.214 -9.801 Variance 0.858 0.941 0.782 0.727 0.671 0.697 0.657 0.771 0.669 0.735 0.584 Skewness 0.434*** 0.259 0.514*** 0.414** -0.102 0.350** 0.082 0.232 0.197 0.383** 0.423** Kurtosis 0.457 0.736** 0.829** 1.250*** -0.154 0.483 -0.168 0.298 0.446 1.044** 0.677* JB 8.861** 7.464** 16.055*** 20.692*** 0.605 6.651** 0.506 2.802 3.259 15.424*** 10.800*** ERS -1.991** -2.219** -3.085*** -2.052** -1.298 -1.664* -0.697 -2.713*** -2.304** -2.367** -2.565** Q(20) 150.01 192.65 76.788* 61.664* 70.390* 49.371* 76.333* 71.481* 51.421* 63.596* 18.901* Q2(20) 129.28 179.59 73.303* 67.099* 58.150* 46.590* 77.409* 70.058* 42.314* 56.189* 16.606* LM(20) 35.265*** 31.601*** 22.115*** 17.969** 25.254*** 15.465 27.585** 17.466** 11.102 16.367* 15.924* Auto 1 0.545 0.498 0.416 0.542 0.555 0.413 0.555 0.402 0.594 0.241 Bankex 0.545 1 0.519 0.397 0.5 0.5 0.378 0.607 0.499 0.805 0.339 Cap 0.498 0.519 1 0.517 0.464 0.561 0.557 0.62 0.475 0.621 0.257 CD 0.416 0.397 0.517 1 0.316 0.397 0.38 0.383 0.391 0.449 0.246 Metal 0.542 0.5 0.464 0.316 1 0.464 0.45 0.556 0.406 0.509 0.217 OilGas 0.555 0.5 0.561 0.397 0.464 1 0.578 0.774 0.475 0.605 0.214 Power 0.413 0.378 0.557 0.38 0.45 0.578 1 0.688 0.434 0.494 0.213 PSU 0.555 0.607 0.62 0.383 0.556 0.774 0.688 1 0.539 0.653 0.245 Reality 0.402 0.499 0.475 0.391 0.406 0.475 0.434 0.539 1 0.491 0.227 BSE 0.594 0.805 0.621 0.449 0.509 0.605 0.494 0.653 0.491 1 0.386 Tech 0.241 0.339 0.257 0.246 0.217 0.214 0.213 0.245 0.227 0.386 1 Table 5. Data Post – COVID-19 Table 5 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech Mean -9.261 -9.575 -9.426 -9.514 -8.673 -9.358 -9.49 -9.391 -8.843 -10.214 -9.801 Variance 0.858 0.941 0.782 0.727 0.671 0.697 0.657 0.771 0.669 0.735 0.584 Skewness 0.434*** 0.259 0.514*** 0.414** -0.102 0.350** 0.082 0.232 0.197 0.383** 0.423** Kurtosis 0.457 0.736** 0.829** 1.250*** -0.154 0.483 -0.168 0.298 0.446 1.044** 0.677* JB 8.861** 7.464** 16.055*** 20.692*** 0.605 6.651** 0.506 2.802 3.259 15.424*** 10.800*** ERS -1.991** -2.219** -3.085*** -2.052** -1.298 -1.664* -0.697 -2.713*** -2.304** -2.367** -2.565** Q(20) 150.01 192.65 76.788* 61.664* 70.390* 49.371* 76.333* 71.481* 51.421* 63.596* 18.901* Q2(20) 129.28 179.59 73.303* 67.099* 58.150* 46.590* 77.409* 70.058* 42.314* 56.189* 16.606* LM(20) 35.265*** 31.601*** 22.115*** 17.969** 25.254*** 15.465 27.585*** 17.466** 11.102 16.367* 15.924* Auto 1 0.545 0.498 0.416 0.542 0.555 0.413 0.555 0.402 0.594 0.241 Bankex 0.545 1 0.519 0.397 0.5 0.5 0.378 0.607 0.499 0.805 0.339 CAP 0.498 0.519 1 0.517 0.464 0.561 0.557 0.62 0.475 0.621 0.257 CD 0.416 0.397 0.517 1 0.316 0.397 0.38 0.383 0.391 0.449 0.246 Metal 0.542 0.5 0.464 0.316 1 0.464 0.45 0.556 0.406 0.509 0.217 OilGas 0.555 0.5 0.561 0.397 0.464 1 0.578 0.774 0.475 0.605 0.214 Power 0.413 0.378 0.557 0.38 0.45 0.578 1 0.688 0.434 0.494 0.213 PSU 0.555 0.607 0.62 0.383 0.556 0.774 0.688 1 0.539 0.653 0.245 Reality 0.402 0.499 0.475 0.391 0.406 0.475 0.434 0.539 1 0.491 0.227 BSE 0.594 0.805 0.621 0.449 0.509 0.605 0.494 0.653 0.491 1 0.386 Tech 0.241 0.339 0.257 0.246 0.217 0.214 0.213 0.245 0.227 0.386 1 Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 172 NET Spillover The column and row (off-diagonal) summation presents "CTO" and "From" directional spillovers. The net spillover gives the direction of volatility spillover. The row total, which is the direction from others to the sectors volatility spillover, is a collection of volatility spillover from each sector contribution. When each sector row sum is subtracted from the column total, we get the net flow of volatility spillover. A positive sign means that the sector is the net provider of the volatility spillover. While negative net volatility spillover would mean that sector is the receiver of volatility spillover. Diebold and Yilmaz Index of Volatility spillover The model's volatility spillover within the model (table 6 & 7) increases from71.29% to 72.59% when we compare the entire sample, and partial sample denoted as S1 and S2 (out of sample and in the sample, respectively). This 1.3% increase in the volatility spillover is substantial, especially if we see that the comparative smaller data set of 221 days has caused this increased volatility spillover. What is noteworthy is that Pre Covid-19 is much calmer comparatively. The volatility spillover is 67.72% for the S2 data set. The graph of DTC (figure 2) also shows the lowermost volatility spillover drops to touch 60%. This does not happen with any other time frame under the study. Table 6. S1 VOLATILITY SPILLOVER Connectedness Diebold and Yilmaz Index Table 6 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech FROM Auto 25.319 9.565 7.364 4.91 7.092 6.477 7.871 8.611 5.786 12.953 4.051 74.681 Bankex 8.469 24.719 8.309 4.666 5.602 5.769 7.752 9.805 5.641 15.359 3.908 75.281 CAP 7.598 9.228 26.073 4.867 5.754 6.43 10.747 9.548 6.865 10.023 2.867 73.927 CD 6.277 7.397 6.755 36.521 4.83 6.376 6.604 6.981 6.984 8.152 3.123 63.479 Metal 8.025 7.315 6.603 4.317 26.488 7.346 9.139 11.706 6.182 9.036 3.844 73.512 OilGas 6.658 6.587 6.427 4.415 6.702 27.514 8.181 13.264 5.446 10.223 4.581 72.486 Power 7.074 7.431 9.556 4.467 7.578 7.55 25.281 12.91 6.771 8.29 3.093 74.719 PSU 6.962 9.101 7.717 4.365 8.425 10.789 11.686 22.767 6.201 8.801 3.184 77.233 Reality 6.806 7.571 7.752 6.329 6.65 6.883 8.664 9.109 28.714 8.325 3.196 71.286 BSE 9.921 13.45 7.965 4.866 6.149 7.762 7.503 8.567 5.024 22.502 6.29 77.498 Tech 6.471 7.439 4.795 4.006 5.834 6.656 6.037 6.355 4.107 12.479 35.821 64.179 CTO 74.261 85.086 73.243 47.208 64.617 72.039 84.185 96.856 59.006 103.642 38.137 798.28 CTI 99.58 109.805 99.316 83.73 91.105 99.553 109.465 119.624 87.72 126.144 73.958 TCI Net SPO -0.42 9.81 -0.68 -16.27 -8.9 -0.45 9.47 19.62 -12.28 26.14 -26.04 72.571 Table 7. S2 VOLATILITY SPILLOVER Connectedness Diebold and Yilmaz Index Table 7 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech FROM Auto 25.579 9.323 7.95 5.297 6.536 5.784 8.048 8.031 6.595 13.085 3.771 74.421 Bankex 8.226 25.052 9.193 4.88 5.599 5.182 7.931 9.814 5.927 15.1 3.096 74.948 CAP 7.598 9.893 26.56 4.168 5.603 5.9 11.231 9.521 7.077 9.747 2.704 73.44 CD 6.31 7.639 6.125 38.343 5.025 5.465 6.63 7.396 7.019 7.47 2.579 61.657 Metal 7.273 7.235 6.901 4.703 26.936 7.016 9.276 11.184 6.801 8.996 3.679 73.064 OilGas 6.052 6.53 6.228 4.299 6.416 29.699 7.982 13.095 5.523 9.896 4.279 70.301 Power 6.947 7.716 10.089 4.556 7.311 7.252 24.931 12.586 7.119 8.391 3.103 75.069 PSU 6.364 9.313 7.92 4.754 7.995 10.515 11.777 23.236 6.606 8.618 2.902 76.764 Reality 6.948 7.423 7.913 6.167 6.876 6.494 8.988 9.324 28.964 7.926 2.977 71.036 BSE 9.988 13.351 8.259 4.766 6.218 7.161 7.713 8.301 5.176 23.15 5.916 76.85 Tech 6.156 6.325 4.886 4.031 5.795 6.185 5.902 5.765 4.253 12.175 38.529 61.471 CTO 71.861 84.748 75.464 47.622 63.374 66.954 85.478 95.016 62.095 101.404 35.006 789.022 CTI 97.44 109.8 102.024 85.964 90.31 96.653 110.408 118.253 91.059 124.554 73.535 TCI Net SPO -2.56 9.8 2.02 -14.04 -9.69 -3.35 10.41 18.25 -8.94 24.55 -26.47 71.729 Table 8. Pre-COVID-19 VOLATILITY SPILLOVER Connectedness Diebold and Yilmaz Index Table 8 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech FROM Auto 31.678 9.748 6.941 5.204 9.368 8.943 4.75 8.251 3.901 10.113 1.105 68.322 Bankex 7.78 28.692 6.784 3.451 6.758 7.394 4.156 10.419 5.778 16.931 1.857 71.308 CAP 7.497 7.264 27.545 7.165 6.365 8.114 8.992 10.196 6.204 9.796 0.861 72.455 CD 6.889 6.57 9.684 40.039 3.95 6.705 5.924 6.107 5.048 7.602 1.481 59.961 Metal 9.274 8.817 7.273 3.579 33.78 6.592 6.492 9.877 5.171 8.481 0.663 66.22 OilGas 8.412 6.477 7.762 4.311 5.577 26.091 9.245 15.747 6.694 9.166 0.519 73.909 Power 5.97 4.159 9.469 4.069 6.211 10.732 30.675 14.704 6.081 7.145 0.784 69.325 PSU 7.265 8.894 8.5 3.43 7.244 13.648 10.997 23.46 6.339 9.358 0.863 76.54 Reality 5.102 9.606 6.915 5.085 5.582 7.167 7.089 10.467 34.518 7.687 0.783 65.482 BSE 8.357 15.876 8.317 4.322 6.543 8.957 6.181 10.309 5.103 23.45 2.585 76.55 Tech 4.295 7.059 4.763 4.637 4.714 2.111 2.993 3.629 3.205 7.401 55.191 44.809 CTO 70.841 84.469 76.408 45.254 62.314 80.363 66.819 99.705 53.525 93.682 11.501 744.881 CIO 102.518 113.161 103.952 85.293 96.093 106.454 97.494 123.165 88.043 117.132 66.693 TCI Net SPO 2.518 13.161 3.952 -14.707 -3.907 6.454 -2.506 23.165 -11.957 17.132 -33.307 67.716 Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 173 Table 9. Post-COVID-19 VOLATILITY SPILLOVER Connectedness Diebold and Yilmaz Index Table 9 Auto Bankex CAP CD Metal OilGas Power PSU Reality BSE Tech FROM Auto 17.731 9.225 9.687 8.004 7.042 8.752 6.055 8.09 4.815 12.176 8.422 82.269 Bankex 7.368 19.278 9.722 6.946 5.228 8.163 5.198 8.207 4.67 15.057 10.162 80.722 CAP 8.6 8.796 18.661 7.725 6.928 8.5 6.292 9.657 5.461 11.806 7.575 81.339 CD 8.882 8.593 9.318 17.659 6.673 8.599 5.63 8.196 5.752 11.249 9.449 82.341 Metal 8.382 6.479 9.219 6.624 17.781 9.281 7.292 10.7 4.697 10.98 8.566 82.219 OilGas 7.857 7.388 9.351 6.422 7.313 16.613 6.596 11.666 5.312 12.056 9.427 83.387 Power 7.763 6.133 9.677 5.989 7.212 8.712 21.133 12.978 5.378 9.086 5.94 78.867 PSU 7.147 7.197 10.307 6.162 8.401 11.404 9.103 16.925 5.351 10.794 7.211 83.075 Reality 7.642 8.311 10.007 6.889 5.301 8.285 6.235 7.826 20.972 10.657 7.874 79.028 BSE 8.833 11.879 9.668 7.311 6.491 9.626 5.722 8.969 4.789 16.452 10.26 83.548 Tech 6.397 10.252 8.59 6.407 4.787 9.568 5.186 7.941 4.327 13.444 23.1 76.9 CTO 78.871 84.252 95.545 68.479 65.375 90.889 63.309 94.23 50.552 117.304 84.886 893.694 CTI 96.602 103.531 114.207 86.139 83.155 107.502 84.443 111.155 71.524 133.756 107.986 TCI Net SPO -3.398 3.531 14.207 -13.861 -16.845 7.502 -15.557 11.155 -28.476 33.756 7.986 81.245 The major jolt of COVID-19 sets rolling a very volatile period, but the lockdown announcement beginning with significant uncertainty shows how volatility spillover touches an all-time high of 81.25% within the model (table 9). The S1, S2, and Pre Covid data set should adhere to the volatility spillover sequence built on historical data. An out-of-sample data should confirm the concurrence of the robustness, which is the post-COVID-19 data set. Hence, it can show if the volatility spillover remains the same or changes. The S2 Pre Covid-19 are identical, showing the similarity in the volatility spillover profile within these two data sets. The sectors receiving the volatility spillover are ranked for better comparison. The rankings can show the sequencing of the volatility spillover changes or not. Does the COVID-19 upsurge the volatility spillover or not is hence answered. The section named "CTO," the first part of table 10 (summarized from table 6 to 9), shows the sector-wise volatility spillover transmission to other sectors. The number in each column shows how much volatility spillover has been induced by that sector. In this part of table 10, we can compare the S2 and Pre-COVID-19 periods. Out of eleven sectors, five (45% of total), the ranking as volatility spillover dispenser remains the same. These sectors are Capital Goods, Auto, Metal, Consumer and Technology, ranking 5, 6, 8, 10 and 11. Table 10 shows the “FROM.” This column shows the volatility spillover received by each sector from others. Out of eleven sectors, four maintain their ranking as the volatility spillover receivers. These sectors are BSE, PUS, Consumer, and Technology, with ranks of 1,2,10, and 11. (36% of all the sectors). The out of sample confirms the validity of the significant volatility spillover movement within the model. The Net Spillover adds the directional explanation to the volatility spillover. The positive sign shows the variable as the transmitter of the volatility spillover, while the negative sign shows that the variable is the receiver of the volatility spillover. This section shows that 36% of the sectors maintain their ranking as either receivers or dispensers of the volatility spillover. Capital Goods remain the volatility transmitter with value of 2, increasing marginally to 3.9%. The consumer sector maintains its position with a negative sign volatility spillover of -14.0% to -14.7%. Technology increases from -26.46 to -33.30 while it maintains its overall ranking of the eleventh position. A noteworthy change is only in the Auto sector, which changes from negative to positive but maintains its sixth position. Has the COVID-19 altered the volatility spillover sequence? The fourth data set of Post should have the exact nature as the S2, Pre COVID-19 data set to answer the question. Change in the ranking of one position up or down can be seen in the data sets S2 and Pre. Those sectors changing with position one rank down or up can be ignored, as the DYI table of (table 10) in the S2 and Pre COVID-19. When we compare the Post or COVID-19 period with the Pre, we see a significant upsurge in the ranking sequence change. In the "CTO" section, we can see that seven sectors change their ranking by two to as many as six positions. The pattern is also seen in the "FROM" section, where six sectors show similar rank movement patterns. The "Net" volatility spillover section shows the change in eight sectors. Table 10. The Ranking Based on the VOLATILITY SPILLOVER Tables Table 10 Volatility Spillover Values % Ranks of Volatility Spillover To S1 S2 PreCOVID PostCOVID S1 S2 PreCOVID PostCOVID Auto 74.261 71.861 70.841 78.871 5 6 6 7 Bankex 85.086 84.748 84.469 84.252 3 4 3 6 BSE 103.642 101.404 93.682 117.304 1 1 2 1 CAP 73.243 75.464 76.408 95.545 6 5 5 2 CD 47.208 47.622 45.254 68.479 10 10 10 8 Metal 64.617 63.374 62.314 65.375 8 8 8 9 OilGas 72.039 66.954 80.363 90.889 7 7 4 4 Power 84.185 85.478 66.819 63.309 4 3 7 10 PSU 96.856 95.016 99.705 94.23 2 2 1 3 Reality 59.006 62.095 53.525 50.552 9 9 9 11 Tech 38.137 35.006 11.501 84.886 11 11 11 5 Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 174 From S1 S2 PreCOVID PostCOVID S1 S2 PreCOVID PostCOVID Auto 74.681 74.421 68.322 82.269 5 5 7 5 Bankex 75.281 74.948 71.308 80.722 3 4 5 8 BSE 77.498 76.85 76.55 83.548 1 1 1 1 CAP 73.927 73.44 72.455 81.339 6 6 4 7 CD 63.479 61.657 59.961 82.341 11 10 10 4 Metal 73.512 73.064 66.22 82.219 7 7 8 6 OilGas 72.486 70.301 73.909 83.387 8 9 3 2 Power 74.719 75.069 69.325 78.867 4 3 6 10 PSU 77.233 76.764 76.54 83.075 2 2 2 3 Reality 71.286 71.036 65.482 79.028 9 8 9 9 Tech 64.179 61.471 44.809 76.9 10 11 11 11 NET S1 S2 PreCOVID PostCOVID S1 S2 PreCOVID PostCOVID Auto -0.42 -2.56 2.518 -3.398 5 6 6 7 Bankex 9.81 9.8 13.161 3.531 3 4 3 6 BSE 26.14 24.554 17.132 33.756 1 1 2 1 CAP -0.68 2.024 3.952 14.207 7 5 5 2 CD -16.27 -14.036 -14.707 -13.861 10 10 10 8 Metal -8.9 -9.69 -3.907 -16.845 8 9 8 10 OilGas -0.45 -3.347 6.454 7.502 6 7 4 5 Power 9.47 10.408 -2.506 -15.557 4 3 7 9 PSU 19.62 18.253 23.165 11.155 2 2 1 3 Reality -12.28 -8.941 -11.957 -28.476 9 8 9 11 Tech -26.04 -26.465 -33.307 7.986 11 11 11 4 Model 72.571 71.729 67.716 81.245 NA NA NA NA The significant changes take place in the ranking of Technology. The I.T. sector is more immune to the market portfolio (primary Index) and the right candidate for portfolio optimisation. Here the COVID-19 alters the position. Technology becomes the transmitter of the volatility spillover—sign changes to positive. However, the change is not significant enough when we look within sample S1 and compare it with S2. The past performance of the sector volatility spillover reduces the volatility spillover from -26.46 to -26.04. BSE as the primary market index doubles its volatility spillover transmission. 17.13 to 33.75. Nevertheless, the ranking changes from two to one. Other observations: Bankex shows a significant volatility spillover reduction. Capital Goods increased volatility spillover 3.6 times from 3.9 to 14.20. Metal increases the volatility spillover by almost four times. Power volatility spillover increases by 6.2 times from -2.5 to -15.5. PSU reduces the volatility spillover from 23.11 to 11.15. Consumer goods show resilience by maintaining the status of net volatility spillover receiver. What is essential is to witness the peculiarity of the COVID-19 volatility spillover mechanism. The lockdown announcement leads to the closure of manufacturing sectors sending a rippling effect on the ancillaries and the supply chain. The bankex shows very robust resistance to the COVID-19. This can be because of the digitalisation motivated by Demonetization. The event of Demonetization had prepared India by a significant shift of retail banking to a digital platform. The episode of COVID-19 has shown bankex as the most robust investment vehicle. CONCLUSION The analysis shed some critical implications of the nature of the COVID-19 pandemic. The data set of 221 days stands out in its COVID-19 effect implications. For some time, it seemed that in India, the situation had rolled back to pre COVID-19. The financial markets rebounded. The gross change in the sector volatility spillover shows the uniqueness of the shock which hit the financial markets in 2020. Factories are operational for the entire four shifts. However, as seen above, the volatility moved in a more differentiated manner. As the second wave of COVID-19 sets in, this volatility behaviour can help the fund managers, regulators, and business managers to forecast and understand how the volatility will unfold. The lessons must be learned how digitisation in India had been a significant mark of help in helping the displaced labour force through online relief transfer into their account by the Indian government. The “Demonetisation,” which was grossly criticised earlier by many, forced the digital payment to an extent. As seen through the bank index “Bankex”, digitalisation has been the least to dispense the volatility. Technology has changed its profile from a volatility receiver to a volatility spiller. The sector based on foreign clientele needs to relook at the risk factors. The work change from the office to home should have made the sector's response resilient, yet it performed the opposite. The takeaway for the PSU and the Power sector is that they need to reassess the risk mitigation strategies under such COVID-19 type shocks. Author Contributions: Conceptualization, K.S.T.; Data Curation: K.S.T.; Methodology: K.S.T.; Validation: K.S.T.; Visualization: K.S.T.; Formal Analysis: K.S.T.; Investigation: K.S.T.; Resources: K.S.T.; Writing – Original Draft: K.S.T.; Writing – Review & Editing: K.S.T.; Supervision: K.S.T.; Software: K.S.T.; Project Administration: K.S.T.; Funding Acquisition: K.S.T. Authors have read and agreed to the published version of the manuscript. Institutional Review Board Statement: Ethical review and approval were waived for this study, due to that the research does not deal with vulnerable groups or sensitive issues. Funding: The authors received no direct funding for this research. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to restrictions. Conflicts of Interest: The authors declare no conflict of interest. Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 175 REFERENCES Albuquerque, R., Koskinen, Y., Yang, S., & Zhang, C. (2020). Resiliency of Environmental and Social Stocks: An Analysis of the Exogenous COVID-19 Market Crash. The Review of Corporate Finance Studies, 9(3), 593–621. https://doi.org/10.1093/rcfs/cfaa011 Antonakakis, N., Cunado, J., Filis, G., Gabauer, D., & Perez de Gracia, F. (2018). Oil volatility, oil and gas firms and portfolio diversification. Energy Economics, 70, 499–515. https://doi.org/10.1016/j.eneco.2018.01.023 Aslam, F., Mohmand, Y. T., Ferreira, P., Memon, B. A., Khan, M., & Khan, M. (2020). Network analysis of global stock markets at the beginning of the coronavirus disease (Covid-19) outbreak. Borsa Istanbul Review, 20, S49–S61. https://doi.org/10.1016/j.bir.2020.09.003 Bahmani-Oskooee, M., & Saha, S. (2019). Exchange rate risk and commodity trade between U.S. and India: an asymmetry analysis. Journal of the Asia Pacific Economy, 0(0), 1–21. https://doi.org/10.1080/13547860.2019.1701307 Baker, S. R., Bloom, N., Davis, S. J., Kost, K., Sammon, M. C., & Viratyosin, T. (2020a). The Unprecedented Stock Market Impact of COVID-19. In Review of Corporate Finance Studies (Vol. 9, Issue April). Baker, S. R., Bloom, N., Davis, S. J., Kost, K., Sammon, M., & Viratyosin, T. (2020b). The unprecedented stock market reaction to COVID-19. Review of Asset Pricing Studies, 10(4), 742–758. https://doi.org/10.1093/rapstu/raaa008 Baumöhl, E., Bouri, E., Hoang, T.-H.-V., Shahzad, S. J. H., & Výrost, T. (2020). From physical to financial contagion: the COVID-19 pandemic and increasing systemic risk among banks. In Econstor. Kiel, Hamburg: ZBW – Leibniz Information Centre for Economics. Retrieved from http://hdl.handle.net/10419/218944www.econstor.eu Boëlle, P.-Y., Ansart, S., Cori, A., & Valleron, A.-J. (2011). Transmission parameters of the A/H1N1 (2009) influenza virus pandemic: a review. Influenza and Other Respiratory Viruses, 5(5), 306–316. https://doi.org/10.1111/j.1750- 2659.2011.00234.x Bora, Debakshi & Basistha, D. (2020). (PDF) The Outbreak of COVID-19 Pandemic and Its Impact on Stock Market Volatility: Evidence from a Worst-affected Economy. ResearchGate, December 2019, 1–23. https://doi.org/10.21203/rs.3.rs-57471/v1 Carletti, E., Oliviero, T., Pagano, M., Pelizzon, L., & Subrahmanyam, M. G. (2020). The COVID-19 Shock and Equity Shortfall: Firm-Level Evidence from Italy. The Review of Corporate Finance Studies, 9(3), 534–568. https://doi.org/10.1093/rcfs/cfaa014 Chandra, S., & Kassens-Noor, E. (2014). The evolution of pandemic influenza: evidence from India, 1918–19. BMC Infectious Diseases, 14(1), 510. https://doi.org/10.1186/1471-2334-14-510 Chaudhary, M., Sodani, P. R., & Das, S. (2020). Effect of COVID-19 on Economy in India: Some Reflections for Policy and Programme. Journal of Health Management, 22(2), 169–180. https://doi.org/10.1177/0972063420935541 Coefficient of Skewness. (2008). In The Concise Encyclopedia of Statistics (pp. 92–95). Springer New York. https://doi.org/10.1007/978-0-387-32833-1_64 Coibion, O., Gorodnichenko, Y., & Weber, M. (2020). The Cost of the Covid-19 Crisis: Lockdowns, Macroeconomic Expectations, and Consumer Spending. Retrieved from https://www.nber.org/papers/w27141 Covid-19 fallout: How the pandemic displaced millions of migrants - NEWS MAKERS News - Issue Date: Jan 11, 2021. (n.d.). Retrieved from https://www.indiatoday.in/magazine/news-makers/story/20210111-displaced-distressed- 1755084-2021-01-03 Diebold, F. X., & Yilmaz, K. (2012). Better to give than to receive: Predictive directional measurement of volatility spillovers. International Journal of Forecasting, 28(1), 57–66. https://doi.org/10.1016/j.ijforecast.2011.02.006 Ding, D., Guan, C., Chan, C. M. L., & Liu, W. (2020). Building stock market resilience through digital transformation: using Google trends to analyze the impact of COVID-19 pandemic. Frontiers of Business Research in China, 14(1), 21. https://doi.org/10.1186/s11782-020-00089-z Economics, D., & Series, W. P. (2014). Where Have All The Workers Gone? The Puzzle of Declining Labour Intensity in Organized Indian Manufacturing (35/2014; 2014, Issue 35). Retrieved from https://hummedia.manchester.ac.uk/institutes/gdi/publications/workingpapers/depp/depp-wp36.pdf Flatten the Coronavirus Curve at a Lower Cost - WSJ. (n.d.). Retrieved from https://www.wsj.com/articles/flatten-the- coronavirus-curve-at-a-lower-cost-11585067354 Friedman, M. (1970). A Friedman doctrine - The Social Responsibility of Business Is to Increase Its Profits. New York Times Magazine, 6(Newspaper Article), 33,122-124. Retrieved from https://www.nytimes.com/1970/09/13/archives/a-friedman-doctrine-the-social-responsibility-of-business-is-to.html Gabauer, D. (2020). Volatility impulse response analysis for DCC-GARCH models: The role of volatility transmission mechanisms. Journal of Forecasting, 39(5), 788–796. https://doi.org/10.1002/for.2648 Hall, M. (2020). Burning Planet: Climate Fires and Political Flame Wars Rage Press releases World Economic Forum. Retrieved from https://www.weforum.org/press/2020/01/burning-planet-climate-fires-and-political-flame-wars-rage India-Measures in response to COVID-19 - KPMG Global. (n.d.). Retrieved from https://home.kpmg/xx/en/home/insights/2020/04/india-government-and-institution-measures-in-response-to- covid.html India: unemployment rate due to COVID-19 | Statista. (n.d.). Retrieved from https://www.statista.com/statistics/1111487/coronavirus-impact-on-unemployment-rate/ Liew, V. K. S., & Puah, C.H. (2020). Chinese stock market sectoral indices performance in the time of novel coronavirus pandemic. 100414. https://doi.org/10.21203/rs.3.rs-29363/v1 Liu, H., Manzoor, A., Wang, C., Zhang, L., & Manzoor, Z. (2020). The COVID-19 Outbreak and Affected Countries Stock Markets Response. International Journal of Environmental Research and Public Health, 17(8), 2800. Tomar, Indian Journal of Finance and Banking 9(1) (2022), 164-176 176 https://doi.org/10.3390/ijerph17082800 Liu, H. Y., Wang, Y., He, D., & Wang, C. (2020). Short term response of Chinese stock markets to the outbreak of COVID- 19. Applied Economics, 52(53), 5859–5872. https://doi.org/10.1080/00036846.2020.1776837 Loginov, A., & Heywood, M. (2020). On the different impacts of fixed versus floating bid-ask spreads on an automated intraday stock trading. The North American Journal of Economics and Finance, 54, 101247. https://doi.org/10.1016/j.najef.2020.101247 Long, S., Zhang, M., Li, K., & Wu, S. (2021). Do the RMB exchange rate and global commodity prices have asymmetric or symmetric effects on China’s stock prices? Financial Innovation, 7(1). https://doi.org/10.1186/s40854-021-00262- 0 Martini, M., Gazzaniga, V., Bragazzi, N. L., & Barberis, I. (2019). The Spanish Influenza Pandemic: a lesson from history 100 years after 1918. Journal of Preventive Medicine and Hygiene, 60(1), E64–E67. https://doi.org/10.15167/2421- 4248/jpmh2019.60.1.1205 McWilliams, A., & Siegel, D. (2001). Corporate Social Responsibility: A Theory of the Firm Perspective. The Academy of Management Review, 26(1), 117. https://doi.org/10.2307/259398 Parkinson, M. (1980). The Extreme Value Method for Estimating the Variance of the Rate of Return. The Journal of Business, 53(1), 61–65. Retrieved from http://www.jstor.org/stable/2352357 Ramelli, S., & Wagner, A. F. (2020). Feverish stock price reactions to COVID-19. Review of Corporate Finance Studies, 9(3), 622–655. https://doi.org/10.1093/rcfs/cfaa012 Salisu, A. A., Sikiru, A. A., & Vo, X. V. (2020). Pandemics and the emerging stock markets. Borsa Istanbul Review, 20, S40–S48. https://doi.org/10.1016/j.bir.2020.11.004 Sehgal, S., Ahmad, W., & Deisting, F. (2015). An investigation of price discovery and volatility spillovers in India’s foreign exchange market. Journal of Economic Studies, 42(2), 261–284. https://doi.org/10.1108/JES-11-2012-0157 Taubenberger, J. K., & Morens, D. M. (2006). 1918 Influenza: the Mother of All Pandemics. Emerging Infectious Diseases, 12(1), 15–22. https://doi.org/10.3201/eid1209.05-0979 The Spanish flu (1918-20): The global impact of the largest influenza pandemic in history - Our World in Data. (n.d.). Retrieved from https://ourworldindata.org/spanish-flu-largest-influenza-pandemic-in-history Xiao, X., & Huang, J. (2018). Dynamic connectedness of international crude oil prices: The Diebold-Yilmaz approach. Sustainability (Switzerland), 10(9). https://doi.org/10.3390/su10093298 Yousaf, I., & Ali, S. (2020). Discovering interlinkages between major cryptocurrencies using high-frequency data: new evidence from COVID-19 pandemic. Financial Innovation, 6(1), 1–18. https://doi.org/10.1186/s40854-020-00213-1 Zimmer, S. M., & Burke, D. S. (2009). Historical Perspective — Emergence of Influenza A (H1N1) Viruses. New England Journal of Medicine, 361(3), 279–285. https://doi.org/10.1056/NEJMra0904322 Publisher’s Note: CRIBFB stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2022 by the authors. Licensee CRIBFB, USA. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Indian Journal of Finance and Banking (P-ISSN 2574-6081 E-ISSN 2574-609X) by CRIBFB is licensed under a Creative Commons Attribution 4.0 International License. http://creativecommons.org/licenses/by/4.0/) http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/