Economic research specific to the spread of a pandemic is relatively recent compared to other fields in economic research. Among the research papers notable are Meltzer et al. (1999). Their paper focused on the impact of an influenza pandemic in the United States (US). They utilised a mathematical model to determine the cost impact of a pandemic on the economy, focusing on the impact of vaccinations. They discovered that economic costs are high whenever death is involved. Therefore, a policy that prevents death would be better for the economy in the long run, such as vaccinations.

Likewise, Smith et al. (2009) corroborated Meltzer et al. (1999) with their research in the United Kingdom. Their research also uncovered that the economic costs for illnesses would inadvertently increase as the fatality rate increases. Nevertheless, they added to the discussion by indicating that closing schools will increase economic impact.

Some studies have shown that obtaining enough vaccines is more critical to determining the economic impact of the pandemic than the actual pandemic itself. These scholars discovered that essential sectors like food production were less affected by a pandemic while less essential sectors, such as luxury goods, were the most affected (Smith et al., 2011). The research in Turkey by Yoldascan et al. (2010) also confirmed the results of the previous two papers in the United States and the United Kingdom.

On the other hand, Chen et al. (2018) decided to use weekly aggregate stock price indices from 1998 to 2008 to determine the impact of the SARS epidemic on stock market integration. They utilised the extreme value theory and separated the sample countries into three categories, namely the source country (China), highly infected countries (Hong Kong, Taiwan and Singapore) and a minorly infected country (Japan). Their results indicated that the SARS epidemic affected the long- run cointegration of the Chinese market with the other markets. The findings implied that the SARS epidemic weakened market cointegration, and thus, arbitrage profits could be generated via portfolio diversification in the event of severe epidemics. In Malaysia, Ali et al. (2010) provided evidence to indicate that investors had over- responded to the SARS epidemic. Their results showed a tendency for investors in the Malaysian stock market to overreact whenever an economic crisis or extraordinary political event occurs.

COVID-19 Pandemic, Public Health and the Economy

Carlsson-Szlezak et al. (2020) attempted to explain the theoretical effects that COVID-19 could have on the economy. They believed that the pandemic would affect the economy through three primary channels. The first of these effects was wealth effects, in which the increase in household savings rates was due to falling wealth, i.e., job losses. This outcome directly affects consumption and market demand. Second is the direct hit to consumer confidence. As consumers shift towards a more pessimistic view of spending, this affects demand negatively and keeps consumers spending frugally. Lastly, supply- side shocks are caused by disruptions in the supply chain.

Several studies have also attempted to explain the effects of the COVID-19 pandemic on the economy. Zaremba et al. (2020) wrote a paper investigating the impact of non-pharmaceutical policy responses to the pandemic on stock market volatility. The findings suggested that government responses significantly impact stock market volatility. Al-Awadhi et al. (2020) also provided findings suggesting that the continuous increase in daily cases and deaths negatively affected stock market returns. Dietrich et al. (2020) surveyed US households to gauge their expectations about the impact of COVID-19. Their study hypothesised that the short-term economic effect of the pandemic would be governed by public expectation. The findings illustrated a high standard deviation in the received responses, implying that public perception of the pandemic’s economic impact was uncertain. They further argued that utilising monetary policy to revitalise the economy would work in the short run but not in the medium run.

Subsequently, Barro et al. (2020) used country-level regression models to compare the current COVID-19 pandemic with the Spanish flu of the early 1900s. Their estimations identified that the economic decline in gross domestic product (GDP) caused by the Spanish flu showed a similar pattern to the COVID-19 pandemic. Compared to the Spanish flu, the COVID-19 pandemic had a more significant percentage decline, albeit with a notably less mortality shock.

In Egypt, Elgin et al. (2020) attempted to construct an index of the global economic stimulus packages that governments had introduced in response to the COVID-19 pandemic. The paper first quantified the economic policies utilised by national governments throughout the COVID-19 pandemic, then used the Principal Component Analysis (PCA) methodology to construct the index. The construction of this index allowed the paper to aggregate and standardise the governmental responses across countries.

Subsequently, the findings indicated a significant correlation between population characteristics, public health-related variables and economic variables with government-announced economic stimulus packages. For example, it was discovered that countries with a higher median age, a lower number of hospital beds per capita and a higher GDP per capita typically had a higher stimulus package announced by the government. The paper also implied that government responses were motivated to react to the pandemic as a health crisis rather than an economic crisis. Barrot et al. (2020) attempted to determine the sectoral effects of implementing social distancing measures in France. By employing a standard model of production networks, they identified that social distancing measures had resulted in a 5.6 percent drop in the French GDP due to work-from-home enforcement and the shutting down of non-essential economic activity.

In China, it was discovered that COVID-19’s impact on the economy was less due to death, sickness or the time sacrificed in taking care of the ill but rather fear, stigma and discrimination (Gong et al., 2020). The paper debated that keeping the economic impact of the pandemic to a minimum is a delicate balancing act between keeping the economy running and preventing a health crisis. It also distinguished the effects of the pandemic by explaining its effects in isolation at the micro, sectoral and macro levels. On the financial issue, a recent research paper in Turkey analysed the stock index changes during the COVID-19 pandemic. Kartal et al. (2020) utilised the XU100 index in Turkey as their dependent variable while employing the MSCI Emerging Market Index, Volatility Index, oil prices, CDS spreads, Treasury Bond Interest Rates, foreign exchange rate, and other variables as the independent variables. They used a dummy variable to account for the presence of COVID-19.

Locally in Malaysia, Chia et al. (2020) measured the impact of the COVID-19 pandemic on the returns on the Kuala Lumpur Composite

Index (KLCI). Using a simple Ordinary Least Squares (OLS) regression model, they discovered the following: Firstly, the daily new cases had a significant but minimal impact on the KLCI. Secondly, the impact of the pandemic on companies was affected by their capital size. Thirdly, Shariah-compliant indices suffered lower losses than their non-Shariah counterparts. It was also identified that the number of daily deaths had the most insignificant impact on the indices. Further research conducted by Ashraf (2020) on the same topic as Chia et al. (2020) attempted to identify the impact at a country level. In contrast to the previously mentioned paper that focused on Malaysia specifically, Ashraf (2020) utilised data from 64 countries instead to identify the impact of the COVID-19 pandemic. He discovered that at a general level, stock markets responded negatively as the number of confirmed COVID-19 cases increased.

Chia et al. (2020) concluded that the response of stock markets towards the number of deaths due to COVID-19 was weak. Furthermore, the results indicated that the impact of the pandemic reduced over time, determining that early stock market reactions during the pandemic had a more robust response to the increasing number of cases, while the reaction became much weaker as time passed. They thus concluded that the market adjusted to the existence of the pandemic at a quick pace. Keh and Tan (2021), Nugroho and Pertiwi (2021) and Tanveer (2021) also corroborated the above studies by identifying that the increase in COVID-19 cases negatively affected the stock market regardless of country. Notably, Keh and Tan (2021) further examined government responses in Malaysia and how they affected the stock market. The paper discovered that different policies positively and negatively affected the economy and stock market. For example, workplace closures adversely influenced the stock market, while stay- at-home requirements, international travel control and income support had positive impacts.

Baharudin et al. (2021) examined the Malaysian government’s policies and how they affected consumer confidence. They utilised surveys to determine the general response to the Movement Control Order (MCO) to achieve their goal. Their survey results indicated that most respondents were optimistic about MCO compliance, and most supported its implementation. Furthermore, it was noted that most respondents were optimistic about the financial assistance offered by the government.

Song et al. (2021) used event studies to identify the impact of key events in Malaysia and their effects on the stock market during the pandemic. They found that the market initially reacted negatively during the early stages of the MCO, but the reaction turned positive over time. They hypothesised that the shift of reactions was due to low interest rates, a large number of stimulus packages and the resumption of economic activity after the initial stages of the pandemic. The paper also argued that government stimulus packages lost their effectiveness in rallying the stock market during the later stages of the pandemic.

Another notable research paper is Ishak and Jiun (2021), which reported that cyclical industries were more severely affected by COVID-19. Besides, Mubarok and Al Arif (2021) examined the impact of the COVID-19 pandemic on Islamic stocks and discovered that they were negatively affected by the pandemic but on a lower magnitude than their conventional counterparts. Azman et al. (2021) corroborated their results. Zainuddin et al. (2021) identified that the pandemic negatively affected the Malaysian export sector, while Habibullah et al. (2021) found that the pandemic had a negative effect on the labour sector.

Among the literature concerning the economic effects of the pandemic, most studies focus on the economic impact by analysing the gross domestic product (GDP) as well as viewing the pandemic’s effect from a health crisis perspective (Meltzer et al., 1999; Smith et al., 2010; Keogh-Brown et al., 2011; Qiu et al., 2018).

The existing research lacks analysis of the effects of government policies in response to the COVID-19 outbreak. Furthermore, most research had a short period (Lee et al., 2020; Chia et al., 2020), which did not consider the market crashes and rebounds during the entire year in addition to the spike in the number of cases towards the end of 2020. Therefore, this study fills the gap by looking into the changes in factors related to COVID-19, government policies and their impact on the stock index.

Based on an earlier study by Lee et al. (2020) and Chia et al. (2020), they expect an increase in the number of new daily COVID-19 cases will result in a negative drop in the market index. Financial markets are expected to react negatively to the news of increasing COVID-19 cases. Therefore, hypothesis H1 is proposed and assumes that the relationship between the daily new cases of COVID-19 and the stock index is negative.

H1: The daily new COVID-19 cases negatively affect the stock index.

A negative relationship is also expected between the daily new deaths of COVID-19 and the stock index. Financial markets are expected to react negatively to the news of increasing COVID-19 deaths. Therefore, hypothesis H2 is proposed as follows:

H2: The daily new deaths of COVID-19 have a negative relationship with the stock index.

The MCO had a direct effect of reducing the level of economic activities by restricting human movement. Therefore, the convention would imply that the MCO would have an adverse impact on the stock index. Nonetheless, Chia et al. (2020) showed that the MCO had an unexpected positive relationship with the stock indices and hypothesised that it was due to favourable market sentiment. It was explained that implementing the MCO was viewed as a positive effort by the government to control the pandemic and resulted in positive market sentiment. However, it is noted that the paper covered a short sample period of only four months and might not have accounted for the MCO’s effects in the long run. Therefore, hypothesis H3 is proposed as below:

H3: The movement control order negatively affects the stock index.

The CMCO was implemented right after the end of the MCO and marked the start of businesses and economic activity reopening. Nevertheless, a negative sign is expected between the CMCO and the stock indices as economic activity was still restricted during the period. Therefore, hypothesis H4 is proposed as follows:

H4: The conditional movement control order negatively affects the stock index.

The RMCO was implemented after the end of the CMCO and further opened the economy. It is believed that the sign of the RMCO with the

KLCI would be positive as economic activity had begun recovering during this period. Therefore, the following hypothesis H5 is proposed:

H5: The recovery movement control order positively affects the stock index.

The moratorium on loan repayment is expected to affect the stock indices positively. Hypothesis H7 is proposed as it is hypothesised that the moratorium’s implementation created positive sentiments. It can be viewed as an effort by the government to reduce the adverse effects of COVID-19 on the economy.

H6: The moratorium on loan repayment positively affects the stock index.

The 2020 Economic Stimulus Package (ESP), Bantuan Prihatin Nasional (BPN), PENJANA Economic Stimulus Package and KITA PRIHATIN Economic Stimulus Package are all expected to influence the stock indices positively. This expectation assumes that a government-mandated stimulus package will create a positive market sentiment, positively affecting the stock indices. Therefore, hypotheses H7 until H10 are proposed as below:

H7: The economic stimulus package (ESP) positively affects the stock index. H8: Bantuan Prihatin Nasional (BPN) positively affects the stock index. H9: The PENJANA Economic Stimulus Package positively affects the stock index. H10: The KITA PRIHATIN Economic Stimulus Package positively affects the stock index.

to influence the stock indices positively because it is assumed that positive market sentiment will be created from its announcement. Therefore, hypothesis H11 is proposed:

H11: The PRIHATIN SME Economic Stimulus Package positively affects the stock index.

The Malaysian government announced the i-Sinar withdrawal on 21st December 2020, allowing funds to be withdrawn from the Employees

Provident Fund (EPF). The news is expected to influence the stock indices positively due to the assumption that positive market sentiment will be created from its announcement. Therefore, hypothesis H12 is proposed as follows:

H12: i-Sinar EPF withdrawal has a positive sign with the stock index.

KLCIt = Kuala Lumpur Composite Index at time t; NCt = Daily new cases of COVID-19 at time t; NDt = Daily new deaths of COVID-19 at time t; BRENTt = Brent crude oil price at time t; VIXt = CBOE Volatility Index at time t; DJIAt = Dow Jones Industrial Average at time t; MCOt = Movement Control Order at time t; CMCOt = Conditional Movement Control Order at time t; RMCOt = Recovery Movement Control Order at time t; MORATt = Moratorium on loan repayment at time t; ESPt = 2020 Economic Stimulus Package at time t; BPNt = Bantuan Prihatin Nasional at time t; SMEt = PRIHATIN SME Economic Stimulus Package at time t; PENJANAt = PENJANA Economic Stimulus Package at time t; PRIHATINt = KITA PRIHATIN Economic Stimulus Package at time t; ISINARt = i-Sinar EPF withdrawal at time t. This study also utilised three diagnostic checks for the OLS model. The diagnostic checks would test for multicollinearity, autocorrelation and heteroscedasticity. The three diagnostic tests used were: Variance Inflation Factor (VIF) for multicollinearity, Durbin-Watson test for autocorrelation and White’s test for heteroscedasticity.