Multivariate Analysis of COVID-19 on Stock, Commodity & Purchase Manager Indices: A Global Perspective

The whole world is going through an unprecedented outbreak of COVID-19 leading to huge economy disruption across the globe. Moreover, the alarming pace of COVID-19 is quite upsetting to the leading nancial stakeholders as it has resulted in the eeing away of the customers, investors, and foreign trading partners. Resultantly, global markets succumbed leading to erosion of more than the US $6 trillion within just one week in February 2020. During the same week, alone S&P 500 index also experienced a loss of more than $5 trillion value in the US. This manuscript attempts to perform multivariate analysis of the global economy during the COVID-19 period. An empirical evaluation of the effect of containment policies on nancial activity, stock market indices, purchasing manager index and commodity prices are also carried out. The obtained results reveal that the number of lockdown days, overseas travel ban and scal stimulus signicantly inuence the economic activity.


Introduction
During this challenging and unprecedented time of COVID-19, the prime concern for each nation is the maintenance of population health. However, it has another signi cant and prolonged impact on the national economy which is currently being overlooked. Hence, during this pandemic, each country is facing two major challenges: a healthcare challenge and an economic challenge. The unmatched spread of COVID-19 has triggered concerns for a severe and extended global recession. IMF (International Monetary Fund) has forecasted a negative per capita GDP growth for more than 170 countries in 2020 due to this pandemic, the most severe since 1930s [1].
The planet has experienced a variety of epidemics, including 1918 Spanish Flu, HIV/AIDS outbreak, SARS, MERS, and Ebola. However, COVID-19 potentially turns out to be the biggest emergency ever recorded in our history. Owing to its sharp spread, the outbreak was declared an international public health emergency on 30th January 2020. The outbreak of COVID-19 has triggered global concern, with 3,181,642 cases and 2,24,301 deaths affecting 215 countries and territories as of 1st May 2020 [2]. The graph shown in Fig.1 represents a constantly rising trend for active COVID-19 cases. The US has highest no. of 1.128 million active cases, the country wise statistics as depicted in Fig.2.
To curtail the progression of the virus, restrictions on the transportation of raw materials and nished goods across states are imposed. The Governments also declared nationwide lockdown that projected a signi cant harm to the economy. There are millions of jobs and livelihoods on the stake. Such an uncertain scenario disturbed the supply and distribution chains in almost all sectors. The extent of this nancial impact is still unpredictable and would depend on the nature and severity of the health crisis, the length of the lockdown and the way the situation unfolds over time. leading to a regional nancial crisis and recession in Asia [10]. The global recession of 2008 was triggered by low in ation [11]. The most recent 2010 recession in Greece was provoked by the effects of the global nancial crisis, systemic aws in the Greek micro and macro economy [12]. Yet, the coronavirus pandemic could trigger a new kind of recession, for several reasons. First, most of the past recessions affected only the single side of the supply-demand chain but this COVID-19 has impacted both the chains equally. Secondly, the effect of past recessions was limited to a particular area only, but this has a widespread impact across the globe. The manuscript has been organized as follows. Section 1 gives a brief introduction of the COVID-19 and its association with the global economy. The impact of similar pandemics on economy is discussed in section 2. A section is dedicated to the related work and the proposed empirical model is presented in section 4. Results and discussion are explained in section 5 and nally, the conclusion is given in section 6.

II. IMPACT OF PREVIOUS PANDEMICS ON ECONOMY
Human history starting from pre-historic to the modern era is replete with deadly infectious disease outbreaks. However, over the last two centuries, pandemics are becoming more regular events than ever.
Degradation of natural ecosystems changed land use patterns combined with high population density, faster global travel, economic integration helps in spreading these highly contagious disease outbreaks to newer countries/regions [13] [14]. Resulting in a negative impact on economies will be higher when the pandemic is highly infectious, even if it has lower virulence [15]. Negative impact on socio-economic activities results due to direct and indirect damages wherein indirect damages are attributed to fear-driven behavioral changes in the public.
Spanish Flu (1918), Asian In uenza (1957), Hong Kong In uenza (1968) are recognized as the major In uenza outbreaks of the 20th century [16]. There has been macroeconomic analysis about the impact of these outbreaks on the availability of labor force, manufacturing output, supply, and demand channels. In the study titled "Global Macroeconomic Consequences of Pandemic In uenza Analysis," by Lowy Institute for International Policy, Sydney, Australia suggests that Spanish Flu caused GDP loss of 3 percent in Australia, 15 percent in Canada, 17 percent in the United Kingdom, 11 percent in the United States [17]. According to another study based on the mortality rate of different regions, authors have predicted that 1918 Spanish Flu Pandemic led to an 18% drop in state manufacturing output in the US [18]. Similarly, the 1957 Flu pandemic led to a GDP loss of 3 percent in Canada, Japan, UK, and the United States.
During the rst two decades of the 21st century, outbreaks of SARS in 2003, Swine u in 2009 and MERS of 2012 had a global impact. Economic impact analysis of SARS suggests that countries that were at the epicenter of the outbreak like Hong Kong, China, Singapore lost billions of dollars of their GDP owing to a downward turn in FDI, Export, Tourism etc. [20]. In Swine u in uenza A H1N1 pandemic in 2009, South Korea had 3,082,113 cases, which represents 6.6% of the country's population, reported a direct and indirect socioeconomic loss of US$1.09 billion [21]. MERS epidemic of 2012 which started in the Middle East went on to spread into 22 countries. In 2015 MERS reached South Korea, where the study on the economic impact shows altered consumer spending behavior post-outbreak [22] and tourism industry was reported a loss of US $2.6 billion [23].

II. RELATED STUDIES
The evolution and spread of COVID-19 have badly disrupted the economy of each country across the globe. The global risk factors have increased substantially in response to this pandemic leading to a highly unpredictable and volatile market. Moreover, the spread of this virus is highly uncertain and quite complex to anticipate. Resultantly, various researchers have suggested different approaches in order to evaluate the effect of COVID-19 on nance by exploring different aspects.
Authors in [24] use a technique to understand the evolution of COVID-19 by exploring seven scenarios. The scenario presented here claims that even a controlled outbreak hugely impacts the global economy in minimum time. Additionally, this study concludes that the government of any county plays a crucial role during this critical time. The government must devise short term, and long-term plans to sustain this economic slowdown. Apart from nancial policies, the government also needs to devise economic and effective health-related policies. It helps to minimize the extent of contagion and thus reduces social and economic costs.
During COVID-19, authors in [25] analyzes its effect on the stock price. It notices an exponential rise in the telecom and healthcare industries, while entertainment, energy, and transportation industry experience an unprecedented collapse. Additionally, authors in [26] demonstrate that this has led to unprecedented and signi cant losses in a very short span. It also claims that it may has substantial long-term impacts as well ranging from bulk unemployment to business collapse across the globe. This belief is further strengthened by the work of authors in [27].
The spending and household consumption in response to this epidemic virus are also explored by authors in [27] [28]. According to the authors, the economic impact of COVID-19 is quite underrated due to its comparison with similar crisis. Authors in [27] claim that this uncertain and unpredictable trend due to COVID-19 might lead to a great recession in the history of the global economy. is evaluated to nd the correlation with fresh COVID-19 cases [30]. A detailed description of the methodology followed is given below:

Materials And Methods
A logistics, manufacturing, and so on. The threshold of PMI is 50 units, any value lesser than that re ects recession; higher the value, stronger is the economy. Table I shows the PMI values of the countries since the COVID-19 outbreak.

Results And Discussion
This section presents the results of the empirical model in terms of correlation. The impact of COVID-19 on a stock market index, PMI, commodity market, and SI are presented in the following subsection:

A. Correlation of COVID-19 with Stock Market Indices
The value of the correlation coe cient represents the intensity of association among variables. In this analysis, correlation matrices of S&P 500, FTSE 100, FTSE-MIB, Shanghai, and BSE-Sensex with respect to COVID-19 cases are demonstrated in Fig. 3, respectively. Blue color demonstrates the correlation index.
The darker blue color represents a higher correlation while lighter color demonstrates a weaker correlation. As can be seen from the graphs, Shanghai has the maximum correlation while the USA demonstrates the minimum value of correlation coe cient.
Further, Table III demonstrates the value of the correlation coe cient; negative values represent the negative association between variables, while positive values represent the positive impact of variables on each other. The highest negative value is for the Shanghai index, representing major negative impact on stock markets, whereas value of the correlation coe cient is minimum for USA markets indicating the least impact by COVID-19 spread.

B. Correlation of COVID-19 with PMI
After stringent policies like Nationwide lockdown, business activities were put to a halt, further impacting the industrial sector to an all-time low. The worst-hit pandemic has taken a toll on all industries.
Correlation matrices of PMIs of the countries with respect to COVID-19 cases are demonstrated in Fig. 4, respectively. The USA has shown the darkest value i.e., the highest correlation, while China shows the lightest value explaining the lower correlation between the values. The lowest value of China is explained by the fact that the economy of China is on the path of recovery after the asco of COVID-19. Table IV represents the value of the correlation coe cient.
The PMI index for the selected countries dropped down to below 50; the value for UK and Italy tumbled tõ 30. Table IV shows the impact of restrictions imposed by the virus with respect to PMIs. The results show that the highest value of the correlation coe cient is for the USA points to the sharpest pace of contraction. Further, for China, the coe cient value came out to be positive and re ects the lower value of the correlation indicating that the economy is on the way of growth, which can be seen from the PMI value of above 50 for China.

D. Correlation of COVID-19 with SI
SI can be taken as a measure to track the impact of enforcement of policy measures by the Government on the spread of COVID-19. The response of the Government of different countries has shown a wide variation concerning policy enforcement; some governments quickly intensify initiatives as soon as the epidemic spreads, while in other countries, the rise in intervention rigor tends to lag growth in new cases. Throughout the outbreak period, more stringent policy responses have generally been imposed. The pace at which these initiatives are taken, however, plays a critical role in mitigating the spread. As can be seen from Fig. 5, the SI shows an upward trend with the increased no. of cases, demonstrating that more restrictions were imposed with the spread.

Conclusion
The novel coronavirus outbreak is probably the most important black swan of 2020. Disruption to industrial manufacturing and foreign trade ows and international logistics networks may be beyond estimations because of the drag caused by extended shutdowns in production. Overall, market sentiment is weak, and nancial volatility is on the peak because the outbreak has adversely affected the prospects for nancial recovery. It is far too early and moreover, by its very uncertain nature, it becomes tedious to determine the ultimate impact of COVID-19 on economic activity and commercial revenue. Here, the authors present an empirical model to understand the effect of COVID-19 on the global economy. For the same, the authors evaluate the correlation coe cient of COVID-19 with stock market index, PMI, commodity market, and SI. The obtained correlation coe cient presents the intensity and severity of its impact on the global economy. Also, elongated lockdown is further wrecking its condition. Hence, it is evident that for the quick recovery of the economy, this epidemic should stop at the earliest. Furthermore, economic recovery and prospering also necessitates effective policy implementation by the government. For now, COVID-19 has resulted in declined demand, decreased costs, and disrupted supply chains around the world and thus all industries are currently struggling. However, it should be re ected that our current situation is just that: a brief nightmare that one day will pass.

Declarations
Funding: The research project is not funded by any organization Con icts of interest/Competing interests: Authors do not have any con ict of interest Availability of data and material: The data that support the ndings of this study related to COVID 19 are openly available at https://www.worldometers.info/coronavirus.
The data that support the ndings of this study related to stock market indices are openly available at https://www.moneycontrol.com/markets/global-indices The data that support the ndings of this study related to stringency indices values are openly available in Oxford University Repository at : https://www.bsg.ox.ac.uk/research/research-projects/coronavirus-governmentresponsetracker   Correlation Matrices of PMI with respect to COVID-19 cases Figure 5