The nexus between the oil price, gold price and asean+3’s stock market indexes, and the contribution of the covid 19 pandemic

The stock markets of the ASEAN-5 countries were found to interact dynamically with their respective significant macroeconomic issues including GNP – Gross National Product, Money Supply,

BÁO CÁO TỔNG KẾT

ĐỀ TÀI THAM DỰ CUỘC THI “SINH VIÊN NGHIÊN CỨU KHOA HỌC”

CẤP HỌC VIỆN NGÂN HÀNG NĂM HỌC 2021-2022

TÊN ĐỀ TÀI: THE NEXUS BETWEEN THE OIL PRICE, GOLD PRICE AND ASEAN+3’S STOCK MARKET INDEXES, AND THE CONTRIBUTION OF THE COVID-19 PANDEMIC

LĨNH VỰC: KINH TẾ

CHUYÊN NGÀNH: QUẢN TRỊ TÀI CHÍNH

Tai ngay!!! Ban co the xoa dong chu nay!!! 17014129035131000000

ĐỀ TÀI THAM DỰ CUỘC THI “SINH VIÊN NGHIÊN CỨU KHOA HỌC”

CẤP HỌC VIỆN NGÂN HÀNG NĂM HỌC 2021-2022

TÊN ĐỀ TÀI: THE NEXUS BETWEEN THE OIL PRICE, GOLD PRICE AND ASEAN+3’S STOCK MARKET INDEXES, AND THE CONTRIBUTION OF THE COVID-19 PANDEMIC

LĨNH VỰC: KINH TẾ

CHUYÊN NGÀNH: QUẢN TRỊ TÀI CHÍNH

SINH VIÊN THỰC HIỆN: Phạm Hồng Phúc

Nguyễn Thanh Hằng

Nguyễn Minh Hoàng

Bùi Hà Hạnh Nguyên

Nguyễn Thị Thu Uyên

NGƯỜI HƯỚNG DẪN CHÍNH: GV Ngô Thị Hằng

THÔNG TIN SINH VIÊN THỰC HIỆN ĐỀ TÀI

1 Tên đề tài: THE NEXUS BETWEEN THE OIL PRICE, GOLD PRICE

CONTRIBUTION OF COVID-19 PANDEMIC

2 Lĩnh vực: Kinh tế

3 Chuyên ngành: Quản trị Tài Chính

4 Giáo viên hướng dẫn:

4

08625782

00

phucpham816@gmail.com

6 Lựa chọn đề tài tham dự các cuộc thi (Tích dấu X vào ô lựa chọn)

Ngoài việc tham khảo Sinh viên lựa chọn tham gia cuộc thi, Hội đồng Khoa học của

Học viện sẽ xem xét quyết định lựa chọn đề tài gửi dự thi theo các tiêu chí của các cuộc thi

CONTENTS

CONTENTS i

TABLES ii

ABBREVIATIONS iii

CHAPTER 1 INTRODUCTION 1

1.1 Rationale and Research Questions 1

1.2 Organization of The Research 4

CHAPTER 2 THEORETICAL FRAMEWORK 5

2.1 Stock Market and Its Relationship with World Oil Price 5

2.2 Stock Market and Its Relationship with World Gold Price 5

2.3 Stock Market and Its Relationship with Other Control Variables 6

CHAPTER 3 LITERATURE REVIEW 10

3.1 International Literature 10

3.2 ASEAN Literature 13

CHAPTER 4 EMPIRICAL FRAMEWORK 16

4.1 Data Description 16

4.2 Methodology 17

CHAPTER 5 EMPIRICAL RESULTS 21

5.1 Necessary tests 21

5.2 Regression Results 23

CHAPTER 6 CONCLUSION AND RECOMMENDATIONS 33

REFERENCES 35

APPENDICES 43

TABLES

Table 0 Summary Table of the Dataset 16

Table 1 Unit Root Test for Time Series: Whole Period (January 2010 - December 2021) 21

Table 2 ARDL model: Whole period (January 2010-December 2021) 23

Table 3 Impulse response: Whole period (Jan 2010-Dec 2021) 27

Table 4 Variance Decomposition: Whole period (January 2010-December 2021) 28

Table 5 Variance Decomposition: Before COVID (January 2010-February 2020) 30

Table 6 Variance Decomposition: During COVID (March 2020-December 2021) 30

ABBREVIATIONS

CPI Consumer Price Index

PR Central Bank Policy

ARDL Autoregression Distributed Lag ADF Augmented Dickey-Fuller Test VECM Vector Error Correction Model VAR Vector Autoregressive Model

CHAPTER 1 INTRODUCTION

1.1 Rationale and Research Questions

Gold is a precious metal that acts as both an industrial commodity and a monetary

or investment asset, used for the purpose of hoarding or exchanging Gold is a liquid, counter-cyclical asset that can help investors accomplish their basic objectives of safety, liquidity, and return by acting as a long-term store of value (Yousef and Shehadeh, 2020) Besides, gold is also known as a leading indicator reflecting the level of inflation, and is always considered as a hedge against inflation, diversifying risks when a financial crisis occurs (Gokmenoglu and Fazlollahi, 2015) When there are fluctuations in the economy, the risk of investing in channels increases, investors will tend to look for gold Gold is also kept in significant quantities by central banks and international financial organizations for diversification and economic security purposes

Oil price can be considered as an indicator of volatility, because fluctuations in world oil price are influenced by the unexpected changes in supply and demand, leading to fluctuations of the exchange rate of an oil importing economy Understanding the volatility

of crude oil prices is critical because it can affect many sectors of the economy and contribute to economic instability in both exporting and importing countries (Gokmenoglu and Fazlollahi, 2015) In detail, most businesses use oil as an input fuel, therefore as the price of oil swings around the world, it affects firms' input costs, causing performance to fluctuate This will have a direct impact on company stock indexes (Nguyen, 2016)

Instead of choosing to invest safely in gold and other precious metals, the emergence

of the stock market offers diversification in the portfolio investment, which has greater appeal to investors With a similar amount of money, investors can choose to invest in businesses that perform well, bring higher returns on investment, but also are possibly exposed to higher risks in comparison to traditional investments including gold

Gold, Oil and Stock Market not only provide investors with different investment channels, but also reflect and simultaneously direct economic activities through their markets’ transaction volume and prices The movements of oil price, gold price and stock

price coupled with their interactions serve as important economic indicators for policymakers in regulating the economy As a result, the relationship between gold price, oil price and stock market has always been a topic of interest to researchers and policymakers

Previous studies provide mixed findings about the relationship between the oil and gold markets with the stock market Specifically, Gokmenoglu and Fazlollahi (2015) conducted research on the Interactions of Gold, Oil, and Stock Market in the United States, mentioning the long run and short run impact of gold and oil price on S&P 500 stock market price index The stock markets of the ASEAN-5 countries were found to interact dynamically with their respective significant macroeconomic issues including GNP – Gross National Product, Money Supply, Consumer Price Index and Money Market Rate, according to Wongbangpo and Sharma (2001)

The long-run relationship is confirmed in Gokmenoglu (2015) indicated that Oil price would lead to a decrease 18% in stock market in long-run, Alamgir and Amin (2021) found the positive relationship between World Oil Price and Stock Market with the evidence from South Asia, Nguyen, Nguyen & Ta (2020) shared a similar result, Oil Price positively affect Vietnam Stock Market Whereas short-term interaction is explored in other previous studies including Nguyen (2018) showed that in short-run, Gold Price produced

no effect on Vietnam Stock Market Index, Tursoy (2017) demonstrated that Turkey experienced a negative relationship between gold price and stock price, and positive relationship between crude oil and stock prices

Besides, the emergence of the unprecedented COVID-19 pandemic, affecting and changing various economies across the globe, has been suspected to distort the relationship between oil, gold and stock market The research revolves around focusing on studying the relationship of macro variables, such as wealth or gold to the stock market, but focusing on developed economies In particular, Singh et al (2021) found the negative relationship between the exchange rates, stock market return and the number of COVID – 19 infections

in G7 countries Pak et al (2020) also agreed with Singh’s results, by conducting research

on the impact of the number of infection cases on the stock market in US, Italy, Spain,

Germany, France, Iran and South Korea and revealed that the number of infection cases and stock market indexes experienced a negative interaction In Asia, Arisandhi and Robiyanto (2022) studied the relationship only of gold price and currency rates on the stock market in 5 countries: Thailand, Singapore, Malaysia, Philippines and Indonesia, found that exchange rate was produced a positive correlation with stock price, and be considered as a better investment in compared with gold

To be the best knowledge of the authors, there is no prior literature covering the relationship between oil, gold, and stock markets for ASEAN, and even ASEAN+3 markets Moreover, a limited number of recent studies have just discovered the impact of the COVID-19 epidemic on the stock market, leaving the contribution of COVID-19 pandemic to the aforementioned relationship open for further discussion

Therefore, we decided to carry the study entitled “The nexus between the oil price, gold price and ASEAN+3’s Stock Market Indexes, and the contribution of the COVID-19 pandemic” The novelty of the study includes: (i) the study, by deploying various

appropriate empirical models (VAR – Vector Auto Regression, VECM – Vector Error Correction Model, and ARDL – Auto Regressive Distributed Lag) into scrutinizing the relationship between the two commodities markets with the stock market, provide reliable findings to enrich the research field of finance and economics; (ii) the study makes an attempt to investigate the impact of the pandemic on the relationship; (iii) the dataset combines both macro variables with oil and gold prices to bring about the most objective result

To successfully achieve those contributions, the study focuses on shedding lights on following research questions:

(i) Do world oil price and gold price have either a short run or long run relationship with the stock market of every single country in ASEAN+3?

(ii) How are these relationships different among countries?

(iii) Does COVID-19 have any impact on these relationships ?

1.2 Organization of The Research

This research paper includes six chapters Chapter 1 is the introduction which consists of three main parts, namely the rationale; research objectives, questions, and scope; organization of the study Chapter 2 presents the theoretical framework related to the relationships between the Stock Market and the other five macroeconomic variables which are Oil Price, Gold Price, Consumer Price Index, Central Bank Policy Rate, and Exchange Rate Chapter 3 reveals the international and ASEAN literature review based on the previous findings of several authors Chapter 4 names an empirical framework that gives the research data description and methodology Chapter 5 presents the empirical results combined with the analysis Chapter 6, the final part, summarizes the main outcomes withdrawn in this paper the base, the policy recommendations, and indicates the study limitations

CHAPTER 2 THEORETICAL FRAMEWORK

2.1 Stock Market and Its Relationship with World Oil Price

Oil has been the cornerstone of the economy in this modern world (Huang et al., 1995) It provides energy to the power industry, creates heat, generates electricity and extracts fuel for running vehicles and planes Moreover, it is also essential for the manufacturing of everyday life necessities including plastics, fertilizers, paints, medicines and so on This demonstrates that oil can be widely used as an indicator to evaluate economic stability since the dependence of production on oil is ineluctable Higher oil prices implicates higher production cost (Gisser and Goodwin, 1986), entailing the rise in price of products and services, the corollary is that input costs will also go up which then hurt the businesses’ financial performance

The definition of stock market, which is a gathering of transactions in which investors can buy and sell shares of any enterprises and other securities, indicates that the stock market index is determined by the investors' buying and selling decisions (Chen, 2022) These investors’ decisions rely on their expectation or judgment on that company’s performance Following this logic, during the period of high oil price, the investors themselves perceive it as a risky and not suitable situation to make investment as they understand how oil price could impact on companies’ robust growth Besides, they also have to tighten consumption expenditure to afford higher energy sources, which impedes their decision of consuming products and investing in the stock market All of these explain conventional beliefs that oil's volatility inversely correlates with the stock market index However, this relationship is only right for nations having production activities that depend much on oil-importing In contrast, the oil-exporting countries definitely bear fruits from selling oil at a higher price, which consequently raises government spending and investing activities, then intensifies the production (Filis, Degiannakis, & Floros, 2011) For such countries, the stock market responds positively to the change in the oil price

2.2 Stock Market and Its Relationship with World Gold Price

For a long time, gold has been commonly known as either a shield against inflation

or a safe investment channel because of its high liquidity and stable supply (Gokmenoglu and Fazlollahi, 2015) Unlike paper money that is printed easily, gold mining only makes a trivial adjustment to its supply over time, therefore, its value is not diminished much Even though gold and the stock market are not directly correlated, throughout history, gold price

is observed to surge during market wobbles (Mani, 2019) Whenever there are factors such

as high inflation, increase in interest rate, fall in GDP and so on that stagger the economy, the stock market will be more likely to become less attractive for investors Instead, they would prefer gold investment to secure their money due to its less varying value over the course of time Ciner et al.(2013) wrote that gold can be utilized to diversify the portfolio

as it has low association with other assets Central banks also preserve gold for the same purpose and to buffer the economic downturn (Kaufmann and Winters, 1989) Since the demand for gold during this time spikes, followed by its higher price This somehow explains what happened in the past when a positive change in the gold price indicates a decline in the stock market In other words, gold and the stock market price normally establish an inverse relationship

2.3 Stock Market and Its Relationship with Other Control Variables

Other than the gold price and the oil price, some of other macroeconomic indicators are expected to induce some impact on the movement of the stock market such as Jamaludin

et al (2017) studies on the relationship between macroeconomics variables including inflation, money supply, exchange rate and stock market in selected ASEAN countries, Humpe and McMillan (2020) examine the long term correlation between interest rate, consumer price index, industrial production and stock price from G7 countries, so on Therefore, to ensure the fitness of the empirical models presented beneath, this study incorporates some of the key macroeconomic indicators together with the gold and oil price

to efficiently observe the movement of the stock markets and derive reliable findings and discussion

Central Bank Policy Rate

Interest rate refers to the cost of using others’ money, which can be understood as the compensation for the risk of default and lending service (Hall, 2022) Explanation for this is that the creditors bear the risk of not getting back the loan, hence, interest rate provides an indemnity for that possible loss Along with the default risk is the inflation risk The lenders have to suffer from the decline in the purchasing power of the money that they loan at the present Because the price of goods and services may go up over time, a dollar today can buy more products than a dollar in the future Therefore, interest rate is the expense that borrowers pay to acquire the ability of using the present value of others’ money rather than accelerating money for years

Interest rate is the central bank policy rate which is imposed on commercial banks’ borrowings directly from the central bank in each country Interest rate is one of the main concerns of investors when it comes to the stock market investing for a reason that it can have an immediate impact on stock price In detail, when the government stipulates that the central bank increases the discount rate on commercial banks’ loans to reduce the money supply or slow down the economy’s growth, financial institutions are charged more to borrow As a consequence, they will inflate the rate imposed on the public's borrowings In terms of individuals, they will have less money to spend on their consuming goods and services This means that businesses can sell less products to customers, which in turn diminishes their revenue At the same time, higher interest rate also deters customers from borrowing money that may have been used to invest in the stock market Both of these lead

to a decrease in the company's stock price In terms of corporations, as the interest rate rises, enterprises have to pay more for not only their new loans but also their existing debts, which make them less profitable followed by the decline in the company’s stock price Inversely, when the government enacted the expansionary fiscal policy requiring the central bank to loosen the discount rate to boost economic growth In this case, both individual and corporation customers reap the benefit of this policy because they can exploit the lower interest rate to borrow more and take advantage of these loans’ present value Individual borrowers will choose to spend more on goods and services to facilitate their life and for the ones who have abundant money, they will pour it into the investment channels like the stock market Meanwhile, businesses will employ their cheaper debt to stimulate merger &

acquisition activities as its operation system, which promises a flourishing in the future cash flow Therefore, either customers are more likely to allocate their money for investing or companies can enhance their business performance, the surge in the interest rate results in higher stock prices

CPI

CPI stands for consumer price index which calculates by taking the total price of market basket in a given year divided by the total price of market basket in the base year then multiplied by 100 (Fernando, 2022) The market basket here consists of all goods and services exchanged in a particular economy which are separated into 8 prime groups: housing, apparel, food & beverages, transportation, education and communication, recreation, medical care and others CPI is a prominent assessment of inflation (Walh, 1982; Bryan & Cechetti, 1993; Shaban et al., 2019) or deflation because the variation in CPI is the expression of the change in price of goods and services over years

An increase in CPI can be interpreted as a sign of inflation, in contrast, a drop in CPI indicates the deflation When inflation happens, it poses a threat to all businesses’ growth, especially manufacturing companies due to the higher price of the input Not to mention that the government may require the central bank to impose a higher interest rate on borrowers to constrain inflation, the ripple effect is as stated in the interest rate part that enterprises will feel less willing to execute outlays which are necessary for their expansion The aftermath is businesses’ profit being squeezed Enough companies being hurt by this will cause the stock market to go down

Exchange Rate

Exchange rate and stock market can have a two-way relationship (Wu, 2005) The common syllogism is that an increase in the domestic stock market gives foreign investors faith about that nation’s robust economic health, hence, attracting them to invest in, consequently, the domestic currency will be appreciated In contrast, when the stock market goes down, the faith vanishes and the foreign investors withdraw their investment away from that country, leaving that country’s currency being undermined So in this case, there

is a positive relationship between the exchange rate and stock market Second relationship

is based on the idea that a firm’s value can be significantly affected by the health of the national currency (Dornbusch & Fischer, 1980) It suggests that when a country’s currency falters, its exported goods become cheaper internationally, which can help to fuel growth and lead to a potential increase in profits for companies whose earnings are export based Therefore, which stock markets have myriads of either listed multinational or export based companies, that stock market index will increase in the opposite direction with the increase

in domestic currency because major of its listed companies’ profits are made under foreign currency The converse case happens for the stock markets in which the stock market contains more import firms than export ones

CHAPTER 3 LITERATURE REVIEW

3.1 International Literature

The relationship between macroeconomic factors and the stock market have been mentioned in many researches, however, the nexus could vary depending on different conditions and times

One of the pioneering studies on the relationship between money supply and stock price belongs to the study of Friedman and Schwartz (1963) The result showed that a rise

in money supply would increase liquidity and credit for stock investors leading to higher stock prices In addition, the research results of Maskay and Chapman (2007) supported Friedman and Schwartz (1963)’s argument in more detail through analyzing the relationship between M2 and S&P 500 index The model also mentioned that there was a positive relationship between changes in these two variables

A research of Jareno, Escribano and Cuenca (2019) focused on correlation analysis related to the relationship between macroeconomics factors, in detail, CPI and the stock market indices of 6 countries such as DAX30 (Germany), IBEX35 (Spain), CAC40 (France), MIB30 (Italy), FTSE100 (UK) and S&P 500 (US) from 2000 to 2014 The study concluded that there was a low correlation between CPI and stock market returns Thus, it could be remarked that the macro variable was independent, which means that the development of CPI did not seem to be related to the evolution of the stock market However, given the opposite results, Subhani et al (2010) studied the relationship between CPI and KSE – 100 (Pakistan) in the period of 5 years from 2004 to 2009 and made a conclusion through the regression model results that there is a negative relationship between CPI and KSE – 100

The nexus between interest rates and stock prices could be observed as the asset – price channel of monetary policy In detail, the policy refers to a decrease of stock prices because interest rates are raised by the central bank Assefa, Esqueda and Mollick (2017) also supported the above theory and concluded that the nexus between interest rates and

stock prices were negative in the developed countries through dynamic panels to examine

in 21 developed countries and developing countries from 1999 to 2013

There are a large number of studies on the linkage between exchange rates and stock prices However, it is not easily identifiable to conclude the relation between the two variables Rose and Jose (2020) studied the relationship between Nifty50 (India) returns and US Dollar – India Rupee Exchange Rates with daily data from 2009 to 2019 to conclude that the relationship among the variables was negative based on the impulse response function Rose and Jose’s conclusion was also supported by Sekmen’s research (2011), which mentioned that there was a negative effect between exchange rate volatility and stock returns in the US, by applying the autoregressive moving average models with the data from 1980 to 2008 But, Tian and Ma (2010), using ARDL cointegration approach, demonstrated that there was a positive correlation between the exchange rate of the renminbi (RMB) against the US dollar and Hong Kong dollar and stock prices (in detail, Shanghai A Share Index)

Besides studying the nexus between macroeconomic factors and the stock market, the inter – relationship between commodity markets (in detail, gold price/ oil price) and financial market (stock market) is also a great interest to many researchers However, the number of studies related to the topic are still limited

Gokmenoglu and Fazlollahi (2015) found the long – run impact of both of gold and oil price changes on S&P 500 stock market price index, in which, the impact of the gold price is higher in both long – run and short – run Arouri, Nguyen and Lahiani (2015) also indicated the significant impact between gold prices and China’s stock prices, by using the VAR – GARCH model, with the data from 2004 to 2011 Papapetrou (2001) studied the relationship between oil price and stock market in Greece, using the VAR approach, observed that oil prices could be considered as an important factor to explain the movements of stock prices In detail, real stock returns would decrease because of a positive oil price

The existence and emergence of the COVID-19 has been incorporated into some of the studies However, they have just investigated the impact of the pandemic to a single macroeconomic indicator rather than the impact of COVID-19 to variations in the relationship between oil, gold, and stock market Specifically, Singh, Bansal, Bhardwaj and Agrawal (2021) studied the relationship between the exchange rates, stock market return and the number of COVID – 19 infections in G7 countries with 2021’s data (from January

to July, 2021), using partial wavelet coherence (PWC), demonstrated that the quantity of COVID – 19 cases affect significantly on G7 countries exchange rate and stock market returns in the long – term

Pak et al (2020) observed that “there were significant negative relationships between the daily number of COVID – 19 cases and various stock indices'' (S&P 500 – US; FTSE

100 – UK; CAC 40 – France; and DAX 30 – Germany) Moreover, the study also mentioned that the world financial and oil markets witnessed a significant decrease (in detail, the oil prices declined more than 65%) because of an increase in the number of COVID – 19 cases globally, especially with the US, Italy, Spain, Germany, France, Iran and South Korea as

of April, 2020 Besides, Alzyadat and Asfoura’s study (2021) also supported Pak et al’s conclusions that there were a negative impact between the number of COVID – 19 daily cases and Saudi Arabia’s stock market (in detail, TASI index of Saudi Arabia) through analyzing by VAR model, the Impulse Response Function (IRF) and ARCH models

Gulyani, Gupta and Singh (2021) studied the nexus between stock market volatility and gold prices during the COVID – 19 pandemic (2020’s datas), using Unit root test, Granger causality test GARCH method and Johansen’s co – integration, indicated that there were no causal relationship between gold prices and stock market prices in the short – run However, given the opposite results, Kumar and Robiyanto (2021) mentioned that gold price returns had a significant impact on both the stock market of India and China by using daily time – series data between 2019 and 2020 and applying the GARCH model and Unit root test

Humpe and Mcmillan (2020) researched the relationship between some macroeconomics factors such as money supply, short and long term interest rate, inflation

and the stock market during 2020 They indicated that there was a positive relation between the stock market and money supply during 2020 (COVID – 19 period), in detail, the rise of money supply attributed to approximately half of the stock market recovery

The relationship between macroeconomic fundamentals, gold, oil, and stock market performance has become a hot topic of debate among financial economists (Ouma and Muriu, 2014), particularly in ASEAN countries, where the stock market and economy are growing at the same time, raising an empirical question about these economic phenomena Several scholars have used a variety of theoretical frameworks to model the influence of gold and oil prices on stock market performance, as well as other macroeconomic issues The majority of research focuses on the stock market's short-term impact on those economic factors

In the period before COVID-19, Wongbangpo and Sharma (2001) found a negative long-run link between stock prices and interest rates in the Philippines, Singapore, and Thailand, but a positive link in Indonesia and Malaysia Jamaludin, Ismail, and Manaf (2017) found that CPI, as a proxy for the inflation rate, has a greater impact on the stock market than the exchange rate in Islamic nations like Singapore, Malaysia, and Indonesia Hussin (2013) and his colleagues in Malaysia discovered that the FBMES (FTSE Bursa Malaysia Emas Shariah Index) and crude oil prices have a statistically significant association, as the Pearson correlation coefficient's significance and big value imply a strong relationship Tangjitprom (2013) discovered that the stock market in Thailand is controlled by the exchange rate and that if the Thai Baht grows in value, so would the stock return Forson and Janrattanagul (2013) discovered that CPI and stock market movements had a negative association Oil and gold prices have both had a detrimental impact on Thailand's stock markets (Raza and partners, 2016) The Philippines stock market follows the same trend, with a short-term negative correlation between gold and the stock market (Alshammari and partners, 2019) Wahyudi (2017), on the other hand, claims that the oil price has a positive impact on the aggregate stock price index, while the exchange rate has

a huge negative impact Using the Johansen Cointegration test for the period January 2000

to November 2019, Selvan and Raj (2019) stated that there is a long-run link between the gold price and the Bombay stock market index (Sensex) in India Masih, Peters, and Mello (2012) found a long-run negative link between the stock market and oil price in South Korea As a result of their findings from our VECM, real stock returns appear to be the primary channel of short-run adjustment to long-run equilibrium, as oil price shocks have two negative effects on firm profitability: it raises production costs for firms, and investors anticipate a decline in profit margins for firms, causing them to make decisions that affect stock market indexes (i.e., selling shares) Chang and Youngho used the limits test to cointegration to study whether the gold price and the Japanese stock price have considerably positive impacts on the Japanese interest rate in Japan, a large oil-consuming and gold-holding country This means that rising gold and stock prices may assist to create expectations of rising inflation over time, leading to an increase in interest rates in the long run Nguyen and Vo (2019) found that the price of oil (OP) has a minor effect on the VN-Index, with the regression coefficient being statistically insignificant, and that the price of gold (GP) has little effect on the VN-Index in the short and medium run in Vietnam Furthermore, long-term inflation as measured by the consumer price index (CPI) has a positive impact on the VN-Index Furthermore, the results of the VECM model show that the exchange rate has a short-term negative impact on the VN-Index

During the COVID-19 period, most research results in this period show that the term relationship between stock markets in Southeast Asian countries and oil prices, gold prices, and other macro variables no longer exists maintained as it was before the global pandemic Using the ADCC-GARCH model, Arisandhi and Robiyanto (2022) found a small but positive relationship between the exchange rate and the stock market in the ASEAN-5 countries of Indonesia, Singapore, Malaysia, and the Philippines, and Thailand, whereas gold has a small but positive relationship with the stock market Variations in the USD/IDR exchange rate had no impact on the CSPI or stock volatility in the Islamic stock market, according to Syahri and Robiyanto (2020), however, changes in the gold price had

long-a significlong-ant positive implong-act Suplong-achok Thlong-akolsri (2020) discovered thlong-at deprecilong-ation of the Thai baht has a negative impact on Thailand's stock market, whilst increases in the global price of gold and crude oil have a negative impact Mukherjee (2022) observed that

during the COVID-19 outbreak, the link between gold, oil, and the stock market shifted Similarly, during the pre-pandemic period, the stock market and gold and oil prices had a long-term relationship However, the link shifted during the pandemic, and the stock price was only influenced positively by current gold prices, and the long-term patterns of these variables are no longer related Prabheesh et al (2020) used the DCC-GARCH model to demonstrate a positive joint movement between oil price and stock market for one of the major oil-importing Asian economies, Korea, during the COVID19 outbreak from 1 January to 8 June 2020 and concluded that declining oil prices describe as a negative signal for the stock exchange In Japan, Narayan, Devpura, and Wang (2021) utilized the GARCH-M (1,1) model to show that the influence of the exchange rate (Yen/USD) on Japanese stock returns was significantly stronger during the COVID-19 era than before During the COVID-19 pandemic in Vietnam, Nguyen and colleagues discovered that the gold price had a negative impact on the VN Index, whereas the oil price has no immediate impact Nonetheless, we have reservations about these findings since observable factors that impact stock market volatility, such as CPI, central bank policy rate, and exchange rate, could lead to empirical models with omitted variables bias

Through a review of prior literature both internationally and in ASEAN+3 cases, it can be seen that empirical studies and empirical conclusions on the relationship between the stock market, oil price, gold price, and other macroeconomic variables in ASEAN+3 are very few Hence, it is necessary to call for further studies to provide reliable findings providing a concrete foundation for producing effective and appropriate policy recommendations in preventing the stock market of ASEAN+3 countries from external shocks driven by international gold and oil markets

CHAPTER 4 EMPIRICAL FRAMEWORK

4.1 Data Description

The dataset utilized in this study is a month – end dataset of different macroeconomic variables suggested by various prior literature (Table 0) including central bank policy rate, exchange rate, world oil price, world gold price, consumer price index, and stock price index for the period from January 2010 to December 2021 from four main sources: International Monetary Fund (IMF); Investing.com; U.S Energy Information Administration (EIA) and S&P global The dataset covers nine countries in research, Malaysia, Indonesia, Vietnam, Thailand, India, Philippines, Japan, Korea, and Singapore

Table 0 Summary Table of the Dataset

Stock Price

Index

VNI SET PSEI KOSPI IDDOW KLCI SGXL NIKKEI NIFTY 50

2010M01 – 2021M12

Monthly Investing Assefa et al (2017),

Rose & Jose (2020), Sekmen (2011), Tian

& Ma (2010), Gokmenoglu &

Fazlollahi (2015), Papapetrou (2001), Humpe and

Mcmillan (2020), Arouri et al (2015), Singh et al (2021)

Note: VNI (Vietnam); SET (Thailand); PSEI (Phillipines); KOSPI (South Korea); IDDOW (Indonesia); KLCI (Malaysia); SGXL (Singapore); NIKKEI (Japan); NIFTY 50 (India); Descriptive statistics of 54 time series of nine countries are tabulated in Appendix 5

During the research process, to serve the purpose of studying the impact of the COVID – 19 pandemic on the relationship of gold, oil and some macro factors (central bank policy rate, exchange rate, consumer price index) to the stock market, the dataset was studied with three samples such as whole period (sample 1: 2010M01 – 2021M12), before COVID – 19 pandemic period (sample 2: 2010M01 – 2020M02) and during COVID – 19 pandemic (sample 3: 2020M03 – 2021M12) – in fact, The World Health Organization (WHO) declared COVID – 19 as a global pandemic on March 11th 2020

The stationary characteristic of the original dataset and its transformation, as shown

in the flowchart in Figure 4.1, explain the specific empirical technique to be used in estimating the equation The initial stage in the regression process is to perform the Augmented Dickey-Fuller Test (ADF) and Phillips-Perron (PP) tests on the entire dataset (all variable series) to see whether there is a unit root process

If the results of the unit root testing on Y series demonstrate that Y series are stationary at level (I(0)), the basic Ordinary Least Square (OLS) will be used for stationary

Y and X series

Source: Ngo Thi Hang (2017), Ngo et al (2020)

Cointegration tests will be performed to assess whether the variables X and Y have

a long-term relationship if the Y series are unit root processes (I(1)) at the level but

stationary (I(0)) at the initial difference Depending on the stationary feature of the X series,

several tests and models will be applied If X and Y (both series are I(1)) are non-stationary

(I(1) at level, but stationary (I(0)) at the initial difference, the Johansen-Cointegration Test

with Error Correction Model (ECM) is used If the X series are stationary at the same level,

the Autoregressive Distributed Lag (ARDL) model will be employed in combination with

the Bounds Test Bounds tests should be used initially to see if two variables are

cointegrated If these two tests reveal a cointegrating vector (a long-run relationship)

between two variables, the next step is to use ECM and ARDL to appropriately assess short-

and long-run pass-through The Vector Autoregressive Model (VAR) will be employed

instead of ECM and ARDL if neither of these tests finds a cointegrating relationship

between two variables

In addition, if cointegration is observed but the regression findings from the ARDL

and ECM models show no meaningful evidence for the projected long-run pass-through,

this thesis will use VAR for variable series The model's stability is determined using the

CUSUM test ECM, ARDL, and VAR models are explained in depth in the following

subsections (4.2.1, 4.2.2, 4.2.3)

4.2.1 Error Correction Model – ECM

The error correction mechanism (ECM) devised by Engle and Granger (1987) is the next stage in our study process The speed at which an endogenous variable returns to equilibrium following any change in exogenous variables is estimated using an error correction model, which demonstrates how to reconcile the short-run behavior of a time series economic variable with its long-run behavior (Pesaran, et al., 2001) The ECM is determined using the following equation:

Where y is the estimated coefficient of ECT-1 (the first lag of error correction term) and ECT are the estimated model's residuals

The lag order is decided using a set of criteria that includes: LR (Sequential modified

LR test statistics); FPE (Final prediction error); AIC (Akaike information criterion); SC (Schwarz information criterion); and HQ (Hannan-Quinn information criterion) The recommended lag order changes depending on the criterion utilized, therefore the lag length that is suggested the most frequently by multiple criteria is chosen for the regressing model

in the end (2)

4.2.2 Autoregressive Distributed Lag – ARDL

A false regression can occur when a nonstationary series is regressed on another nonstationary time series The purpose of the co-integration test is to see if there is a long-run equilibrium connection between variables that are integrated at least in first difference

I (1) If the time series variables are integrated at separate levels (I(0) and I(1)), the ARDL (autoregressive distributed lag) approach may be used to look for co integration

The equation for the ARDL modified model is as follows:

Where Δ is the difference operator, Yt denotes the dependent variable, Xt denotes the independent variables, Zt denotes the variable lags, Z = f(Xt, Yt), t denotes the trend term, and is the disturbance

After that, the model's serial correlation and stability should be verified The null hypothesis for testing co-integration is πyy = πxx = 0 (no co-integration), which is evaluated using the Wald test and should be rejected in terms of the long-run connection between variables Narayan provided the F-statistics key values (2005) The presence of a significant

F test indicates co-integration

4.2.3 Vector Autoregressive Model – VAR

According to Isakova (2008), the VAR model is estimated as follows:

where Y denotes endogenous variables and X denotes exogenous variables, A(L) and B(L) denote coefficient matrices, and u denotes a vector of impulses

In a VAR model, ordering variables play a critical role in the estimation process The most exogenous variables are generally ordered first, followed by the least exogenous variables This paper only looks at VAR's variance decompositions (evaluate the contribution of variables in forecast error of one variable) and impulse response functions (trace the impact of one-time shock in one variable on current and future movements of other variables), which provide information on short-run dynamics or immediate interest rate pass-through, in an attempt to investigate the response of stock price to shocks to oil, gold price through a two-stage mechanism

CHAPTER 5 EMPIRICAL RESULTS

5.1 Necessary tests

Unit Root Test

The empirical results for the variables used in this study are presented in this section EVIEWS 12 software was used to do the data estimations and model tests Except for CPI and the central bank’s policy rate, the logarithmic form of time series variables like Oil Price, Gold Price, Exchange Rate (USD is used as the common denominator), and Stock Price are used to observe the percentage change of series and to apply formal tests of stationarity The Augmented Dickey-Fuller Test and the Phillips-Perron Test are used to verify the time series' stationarity Tables 1, 2 and 3 indicate the results of ADF and PP unit root tests of all series included in 3 different samples

as mentioned earlier: the whole period (from January 2010 to December 2021), the period before COVID (from January 2010 to February 2020), and the period during COVID (from March 2020

to December 2021) However, sections 5.1 to 5.2.3 only focus on interpreting the results in the whole period For the other periods, their tables can be found in the appendix 1

Table 1 Unit Root Test for Time Series: Whole Period (January 2010 - December 2021)

Country Variable Null Hypothesis: Variable has a unit root

Time Series at Level Time Series at First Differences

t-stat p-value t-stat p-value t-stat p-value t-stat p-value

LNER -1.10 0.72 -1.10 0.72 I(1) -12.71 0.00 -12.71 0.00 I(0)

LNIDDOW -2.85 0.05 -2.81 0.06 I(1) -11.41 0.00 -11.55 0.00 I(0)

Japan CPI -2.48 0.12 -2.45 0.13 I(1) -10.07 0.00 -9.99 0.00 I(0)

LNKOSPI -1.46 0.55 -1.46 0.55 I(1) -11.79 0.00 -11.79 0.00 I(0)

Malaysia CPI -3.74 0.01 -2.80 0.06 I(1) -8.78 0.00 -7.69 0.00 I(0)

LNER -2.01 0.28 -2.14 0.23 I(1) -11.13 0.00 -11.11 0.00 I(0)

LNSET -3.25 0.02 -3.27 0.02 I(0) -11.01 0.00 -10.99 0.00 I(0)

Vietnam CPI -2.22 0.20 -1.71 0.42 I(1) -3.29 0.02 -5.57 0.00 I(0)

PR -1.45 0.56 -1.21 0.67 I(1) -4.83 0.00 -7.98 0.00 I(0)

LNER -3.56 0.01 -4.16 0.00 I(0) -12.77 0.00 -12.88 0.00 I(0)

LNSP 0.10 0.96 0.17 0.97 I(1) -11.89 0.00 -11.90 0.00 I(0)

Note: I(1) – Unit Root Process; I(0) – Stationary Process; OP: World oil price; GP: Gold price; PR: Central bank policy

rate, ER: Exchange rate, SP: Stock index; LN – logarith

Source: Authors' estimations

Looking through tables, not all variables are non-stationary at level and stationary at first different As presented in the methodology part, for variables are integrated at different levels (Y series is I(1) (not-stationary at level but stationary at the first difference series while X series could

be I(0) or I(1)), the ARDL model is employed to figure out the long-run relationship Results of unit-root test from Table 1 suggest that there are a total of 6 countries including Vietnam, Singapore, Korean, India, Japan, Indonesia that ARDL can be applied to explore long-run relationships between stock market and oil, gold markets, which is discussed in the below section

5.2 Regression Results

5.2.1 Long-term relationship (ARDL Model)

Table 2 ARDL model: Whole period (January 2010-December 2021)

RELATIONSHIP

Short run Adjustment Speed (ECT)

statistic

F-F-critical 10%

F-critical 5%

F-critical 1%

Source: Authors' estimations

The long-run relationship, between stock price index with the world oil price, world gold price and other control variables for all these 5 countries, namely Vietnam, Korea, Indonesia, India, and Japan in the entire research period (2010M01 – 2021M12) are investigated through using the ARDL model as these variables are integrated at different levels I(0) and I(1) There is a long-run relationship among the variables (Stock Price Index;

LNOP; LNGP; PR; LNER; CPI) because the null hypothesis (H0): “no long-run relationship exists” is rejected as the value of F – statistic is 2.843, 9.196, 2.974, 8.344, 4.521, in these countries respectively, which is higher than the value of F-critical 5%

As can be seen from the table, there are two types of relationships between the stock price index, oil price, and gold price, which are the positive and negative ones Hence, the two groups of countries will be examined separately according to the nature of the relationship The first group will be these countries that witness a negative relationship between the variables, namely Vietnam, Korea, and Indonesia Another will be the ones that demonstrate a positive relationship between these variables, which is India and Japan

Vietnam is an outstanding country in the first group, which shows a negative relationship between both oil price and gold price on the stock market index Through results from the long–run ARDL model coefficients, it is clear that a one percent rise in the oil price (LNOP) and gold price (LNGP) would lead to a decrease of approximately 0.101% and a decline of 0.816% in the stock price index in the long run Since gold is seen as a classic stock market replacement An increase in the price of gold may encourage investors

to withdraw their funds from the stock market, resulting in a drop in the stock index The findings are consistent with those of Korhan and Negar (2015), who discovered a negative link between the stock price index and the gold price In terms of oil price, it has a direct and indirect impact on stock market volatility The direct effect is explained by the fact that oil price volatility creates uncertainty in financial markets, which leads to a drop in stock prices When oil prices rise, the indirect effect is stated as a fall in production output and

an increase in the rate of inflation This has an impact on the economy's macro factors, which in turn has an impact on the stock market According to Jones and Kaul (1996), the price of oil has a negative influence on the stock market Because of high input costs and difficulty selling output items, the company's real cash flow is reduced as a result of oil price volatility Furthermore, the study indicated that oil price is a risk factor for the stock market by employing the arbitrage business model In addition, Jones et al (2004) discovered evidence that high oil prices will raise the economy's production costs, resulting

in lower output and a lower projected return on the stock market

In brief, it could be said that the impact of the global gold and oil price on the movement of Vietnam's stock market is relatively weak The short-term disequilibrium relationship of those three markets is adjusted to the long-run one with the adjustment speed

of only 8.3% which is considerably low as well

The results of Korea and Indonesia also reveal the same pattern One percentage increase in oil price leads to a decline of 0.023% in the stock price index of Korea And one percent rise in the gold price in Indonesia leads to a decrease of 0.504% in the stock price index of this country

For the second group of results, taking India as a representative, the result of the ARDL model reveals a positive relationship between the stock price index, oil price, and gold price It indicates that a one percent increase in oil price and gold price would lead to

a rise of 0.120% and 0.384%, respectively in the stock price index, with the adjustment speed of 61.4% The findings are in line with those of Najaf and Najaf (2016), who looked

at the influence of crude oil on the stock exchange They used data from the previous 15 years and utilized Karl Pearson's coefficient of correlation to arrive at the conclusion that lower crude oil prices had a negative influence on the stock exchange According to the findings, oil is the most important source of energy all over the world They said that oil prices are the most important necessity of every country and that as a result, prices have an impact on the country's performance The findings are consistent with those of Raza et al (2016), who looked at the asymmetric influence of gold, oil, and their associated volatilities

on developing market stock markets Gold prices have a positive impact on stock market values in the growing BRIC nations, according to the empirical findings Japan’s results also showed the same outcome One percent increase in oil price and gold price leads to a rise of 0.107% and 0.485% in the stock price index of this country It supports the conclusions of Ta and partners (2020), Nguyen (2018), Hsing (2011), Kuwornu (2011), and Rahman et al (2009), about the positive relationship between oil prices and stock indexes

For ARDL/VECM models along with VAR approaches, the regression models have been checked for common regression issues including the autocorrelation, heteroskedasticity, and the stability of the model The results of tests, LM and CUSUM