**1. Introduction**

It is expected for someone to think that there is no connection between a pandemic and the energy industry. But when it becomes clear that the pandemic is responsible for uncertainty, the relationship between these two magnitudes acquires a logical basis. Two years after the COVID pandemic appeared in the city of Wuhan in China in 2019 and spread rapidly in Europe and in the USA, it was obvious that consequences were going to be severe for the world economy. The main feature of this pandemic was its rapid and unprecedented negative impact on economic activity and in particular the spread of grea<sup>t</sup> uncertainty worldwide. It was then expected for this uncertainty to combine with financial turmoil pushing companies and individuals in taking precautionary measures. Their first measure was to decrease their spending to be able to face impending difficulties if necessary.

The environment created by the financial hardship but mostly by the daily announcements of deaths and infections naturally had a negative effect on the consumers' psychology. In such an insecure environment, it was rather normal to observe a reduction in demand for oil and therefore a reduction in its price. Speaking with numbers, during the first two years of the pandemic, global electricity demand was decreased by an average of 15% [1] resulted in a downward movement of the prices for crude oil and natural gas.

It should be noted that the COVID-19 pandemic period did not have the same characteristics as other periods of uncertainty which were due to a slowing or overheating of the economy, and thus measures to deal with COVID-19 should be different too. In any case,

**Citation:** Christopoulos, A.G.; Kalantonis, P.; Katsampoxakis, I.; Vergos, K. COVID-19 and the Energy Price Volatility. *Energies* **2021**, *14*, 6496.https://doi.org/10.3390/en14206496

Academic Editor: Carlos Henggeler Antunes

Received: 29 July 2021 Accepted: 4 October 2021 Published: 11 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the focus of these measures was on the restriction of free movement and transportation. In fact, travel bans not only applied between different countries but were mainly imposed within countries themselves. So, it was natural for these restrictions on the movement of people and the transport of goods to cause a significant reduction in the demand for fuel, damaging further the energy sector.

At this point it is useful to mention that global organizations and states acted instantly, adopting measures to limit the effects of the pandemic in the global economy. The experience of the past and the tools that technological transformation provides to the policy-makers make decision framework more effective and also more complex. The main objective is to avoid a new universal quarantine of the population in their homes or even partial lockdowns of economic activities.

The main purpose of this paper is to examine the dramatical impact of the COVID-19 on the energy sector during the pandemic period, providing useful results for policymakers, especially in case of a new wave of infections and deaths in the future.

The rest of this paper is organized as follows. Section 2 discusses the existing literature, emphasizing on the effects of COVID-19 on the energy sector while Section 3 presents the data and hypotheses. Section 4 presents the methodology applied to show the relationship between COVID-19 infections and death announcements with oil price volatility. Section 5 explains the empirical results and Section 6 presents the conclusions and a short discussion for future works.

#### **2. Literature Review**

In the recent literature we found studies that investigated the effects of COVID-19 on specific sectors or economic zones or even the global economy. Most of these studies focus on financial markets and the energy sector, while in several cases, macroeconomic factors are also examined. Most of these studies provide useful conclusions, and the effects of the pandemic are expected to remain at the top of academic interest for the next period. The purpose of these studies is to acquire the necessary knowledge to deal with similar cases in the future, in order to limit the negative effects and to reset the real economy and social life on track as soon as possible.

Ref. [2] examines the linkage of the global oil market with the USA energy stock market using their implied volatility indexes. The main conclusion of this study is the existence of a long-run relationship between oil and stock market implied volatility indexes. In a similar way, ref. [3] studied the dynamic correlation between spot oil price fluctuations and the stock uncertainty index for the USA, Japan, China, and Hong Kong in order to find out whether crude oil can be used as a hedging instrument. According to the applied wavelet coherence analysis, crude oil cannot support hedging on a long run period but it can be a hedging instrument in a state of panic, like the pandemic period.

Ref. [4] applied a heterogeneous autoregressive realized volatility model to examine the predictive power for oil-market volatility using an uncertainty index based on the daily newspaper news for the pandemic period. They found that by incorporating such information in their model, forecast accuracy improves significantly.

Ref. [5] applied a nonlinearity autoregressive distribute lag model to examine the crude oil price fluctuation while they also use an event study model to compare how different types of events affect crude oil price fluctuations. In their effort to combine crude oil price fluctuation with what causes it, a state-space model was applied and evidence of strong correlation between event shocks and event types was found.

Ref. [6] attempted to estimate the out-of-sample predictive power of crude oil price volatility in relation to financial ratios and macroeconomic variables which are commonly used in the literature. Her findings suggested that considerable economic profit is possible based on this model while useful implications are also provided for portfolio optimization and asset allocation. On the other hand, ref. [7] examined for the USA the relation between COVID-19 and oil price volatility, the stock market and the geopolitical risk among others. By applying wavelet approaches they found that COVID-19 effect on geopolitical risk is higher than economic uncertainty in the USA.

Ref. [8] provided a way for increasing energy efficiency and energy saving. They examined the challenges of COVID-19 for the energy sector. In particular, they investigated new practices enforced by the pandemic and the way they affected energy demand and consumption. They found that demand has declined but intensity showed apparent changes as the extra energy used to fight COVID-19 was not negligible for the recovery of the demand for energy, while differences in recovery can also be found between different regions.

Ref. [9] examined the implications of COVID-19 for the sustainable energy transitions. The adopted measures by the states, firms, and individuals have motivated many changes that may influence the sustainable transition of energy. They identified the main impact of lockdown on energy and investigated how economic stimulus packages can shape energy transitions and found that the politics of sustainable energy transitions are at a critical stage.

Ref. [10] examined the risk transmission from the COVID-19 to metals and energy markets and found significant negative volatility transmission from the COVID-19 to gold, palladium, and brent oil markets. According to these results, COVID-19 risk is not transmitted to the industrial metal market but COVID-19 leads to an increase in oil market volatility.

Ref. [11] estimated the price volatility of crude oil and natural gas for the listed firms in the MCX exchange of India. Their results are interesting for policy-makers to assess the appropriate strategy in facing the effects of the pandemic as they find leverage effect of COVID-19 on the price volatility of crude oil but not on the price volatility of natural gas.

Ref. [12] examined the hourly oil price volatility and found a significant increase of volatility in the pandemic period. To achieve that, they built a dataset with hourly oil prices combined with global cases of COVID-19 and deaths and applied an OLS regression model with volatility being one of the proxies of oil price volatility. In addition, ref. [13] attempted to estimate predictors of oil prices and for that he examined the interconnection of oil prices with COVID-19 infections and oil price news. He found that effect on oil prices is more significant when infections exceed the threshold of 84,479, whereas the effect of oil price news conditioned on COVID-19 cases is limited.

Ref. [14] estimated the historical volatility of energy markets during the COVID-19 pandemic period by using infection ratio, economic policy uncertainty index and infectious diseases market volatility. His findings can explain the investors' position in implementing options to protect from risk in the energy market and their willingness to pay excess premium for that.

Ref. [15] investigated the relation between the COVID-19, the crude oil market, and the stock market by observing return and volatility spillover with both a time-domain approach and a frequency dynamics approach. Their analysis showed that spillover return mainly exist in the short term while volatility spillover mainly exists in the long term. They also applied a moving window analysis to conclude that COVID-19 created more risk for investors which resulted in high losses in the short term. It is also interesting that COVID-19 impacts on the volatility of the oil, and stock market was even higher than volatility caused in 2008 by the global financial crisis.

Ref. [16] examined the role of gold as a hedging instrument against crude oil price risks. They applied an asymmetric VARMA-GARCH model to assess the impact of COVID-19 and they found that gold can work as a hedge instrument against oil risks as their results during pandemic show negative coefficient of returns spillovers from gold to oil price returns, meaning that an increase in gold in this period will lead to a less decline in oil price returns. Moreover, volatility spillovers between the gold and oil price returns sugges<sup>t</sup> that significant volatility effects are present.

Ref. [17] examined the predictive power of oil supply, demand, and risk shocks in relation to the realized volatility of the daily oil returns. They applied a heterogeneous autoregressively realized volatility approach and showed that especially financial marketdriven risk shocks can improve the forecasting performance for in and out-of-sample. Their

conclusions offer to investors a valuable way to use traded assets at high frequency in order to monitor oil market volatility.

Ref. [18] emphasized on vaccines by examining the storage conditions based on their thermal load to cool and found that the cold storage of Oxford–AstraZeneca, Janssen COVID-19, and CoronaVac vaccines in Brazil generates 35-times less environmental impact than Pfizer. They also developed an energy index showing that Oxford–AstraZeneca, Janssen COVID-19, and CoronaVac vaccines have 9.34-times higher energy efficiency than Pfizer.

Ref. [19] considered that COVID-19 led to an economic crisis which has changed the social behavior and reduced the industrial activity and the demand for power worldwide. To examine the impact of COVID-19 on power demand, they quantified the country load reduction of COVID-19, based on the active cases and the lockdown period as proxies. They found that in Germany and Great Britain the power demand was reduced while in France the demand was increased for the period outside the lockdown. During the lockdown, France had a much higher reduction than in Germany and Great Britain. However, the effect of COVID-19 on carbon emissions in the power sector was small.

Other studies focused on the impact of COVID-19 pandemic on stock market returns and stock market volatility.

Ref. [20] examined the response of stock market returns from 64 countries to confirmed deaths from COVID-19. His research covers the period January to April 2020 and shows a negative response of stock markets returns to confirmed deaths. Ashraf's research also suggested that negative response was stronger and faster in the first days of confirmed deaths indicating that market response depends on the period of the outbreak.

Ref. [21] investigated the effect of official pandemic announcements on financial markets volatility as expressed by S&P 500 and found that COVID-19 is a significant source of price volatility in the USA financial markets which thereafter affects the global financial cycle.

The existing literature highlights reasonable questions about the impact of COVID-19 on oil price volatility and in this paper we try providing some answers to this issue.

#### **3. Data and Hypotheses**

One question that has caught the interest of the academic community in recent years is the relationship between the pandemic and the oil price volatility. Until 2019 the literature showed that oil prices fluctuate due to the forces of supply and demand. Indeed, several research showed that during normal periods, the demand for oil is shaped by global economic activity, while on the supply side, factors related to technological innovations that improve the oil production process are incorporated.

However, in the last two years, we have observed increasing volatility in the price of oil without any economic event justifying it during the same period. As a result, the academic community has turned its attention to investigating the causes that led to this phenomenon. As it turned out, the intense uncertainty created by the pandemic, first for health reasons and then for the possible effects on the world economy, significantly affected the demand for oil. This demand shock is different from the traditional aggregate demand shock because the decline in consumer confidence is inextricably linked to the fear caused by the virus.

Consequently, one of the main questions raised by the literature is how COVID-19 death announcements and the speed of COVID-19 deaths and infections affect oil price volatility.

In this context, our paper provides answers on three theoretical questions contributing with its findings in the existing literature as follows:

**Hypotheses 1.** *How the announcements of new cases affect the volatility of the oil price?*

**Hypotheses 2.** *How the rate of change of cases affect the volatility of the oil price?*

**Hypotheses 3.** *Do the above influences differ between different geographical areas worldwide?*

The three hypotheses are tested with a general econometric panel model similar to the one proposed by ref. [21] who examined the response of financial market volatility on COVID-19 new cases of infection and the fatality ratio. Yet, this article (ref. [21]) does not examine the volatility of oil prices in relation to the pandemic which is the main scope of our study.

In the empirical analysis we collect daily data for COVID-19 infection announcements and deaths from the World Health Organization. Our daily data also include three crude oil volatility indices, the CBOE 30-day crude oil implied volatility index, the 3-month crude oil implied volatility index, and the Brent 3-month implied volatility index. Further, daily data involve also four market uncertainty indices, namely VIX Volatility Index, VSTOXX Volatility Index, NIKKEI Volatility Index, and CBOE China ETF Volatility Index. Last but not least, daily data concern the Economic Uncertainty index, the Baker, Bloom and Davis index of economic policy uncertainty for Europe which is based on the frequency of newspaper references to policy uncertainty.

In the first stage of analysis, our sample is divided into six main geographical areas, so that the empirical analysis leads to useful conclusions for each area separately and in the second stage of analysis the population of the six geographical areas is an aggregate sample.
