Quantification of Stock Market Co-Movement Between Hong Kong and Singapore

1.     Introduction

Stock market co-movement has been one of the most significant discussions in finance. Generally, co-movement refer to the positive correlation between two or more variables (Barberis, 2002). However, the precise explanation of the positive relationship is complicated to depiction. In finance, the definition of co-movement is the changing of one asset price to another asset price (Baur,2004). Along with the increasingly globalised economy, the degree of stock markets integration has attracted much attention by researchers and investors to analyse the link between one national stock prices with another, fundamentally due to its fact of being related to every aspect of finance such as portfolio diversification, asset allocation and risk management (Hoque, 2007).

In this paper, we will attempt to quantify the stock market co-movement between Hong Kong and Singapore. As two of four “Asian tigers”, both Hong Kong and Singapore are advanced economy entities located in Southeast Asia (Young, 1992). They share considerable similarities in history, culture and financial aspects. Historically, both Singapore and Hong Kong had a past of been colonised by the United Kingdom (Chen,2016). Regarding culture, the majority of residences in Hong Kong and Singapore are ethnic Chinese, which accounts for more than eighty per cent of the total population (Young, 1992).

Economically, the scarcity of natural resources and small-scale landmass prompt Hong Kong and Singapore to rely heavily on international trade (Chen, 2016). Before becoming the world’s financial centres, each is changed the industry structure from labour incentive like clothing to high-tech to finance and banking service in the last decades (Chen, 2016). Moreover, Hong Kong and Singapore have been close business partners in various fields for decades, and they practically behave the same in economic performance and various rankings. For instance, Singaporean leading financial services groups such as DBS and OCBS, continuously expanding their business in Hong Kong in the last ten years (Ngiam, 2013). The GDP per capita for both Singapore and Hong Kong reached more than 40,000 dollars every year (Ngiam, 2013). Based on the global competitiveness report published by International Institute for Management Development (IMD) in 2019, Singapore and Hong Kong are the only two Asian regions in the top 10 of the ranking and won the first and second place respectively.

The paper is structured as follows. Section 2 explores the theoretical framework. Section 3 examines the literature review regarding stock market co-movement. Section 4 and 5 demonstrate data and methodology, severally. Section 6 and 7 presents empirical results and analysis. Finally, conclusion and limitations are illustrated in section 8 and 9.

2. Theoretical Framework

2.1. The Modern portfolio theory

In finance, investment refers to the purchasing of assets that are likely to generate profit in the future (Groppelli and Nikbakht, 2006). The investment portfolio may consist of both tangible assets and intangible assets. The future performance of some intangible assets, such as stocks is unpredictable. The modern portfolio theory explains that after examining a series of assumptions, researchers find a way of balancing risks and returns to maximize investors satisfaction, it is also seen as the efficient frontier (Milne, 2004). The fundamental aim of the investment is to put different kinds of assets that have various risk and return level into a basket to achieve the most considerable profit by taking minimum risks (Groppelli and Nikbakht, 2006).

2.2. Diversification

Diversification is to invest in different assets that are not entirely correlated with one another to mitigate portfolio risks (Milne, 2004). In general, high profit comes with high risks. Fabozzi and Peterson (2003) claim that the portfolio risk cannot be removed entirely in any way as the same effects and innovations will have influences on assets. By investigating the standard deviation of randomly chosen portfolios, Hagin (2004) confirmed that portfolio volatility would decrease along with the increasing number of stocks added into the portfolio.

Diversification can be achieved by enlarging investment locations. Odier and Solnik (1993) find that international investment has significant advantages for investors to decrease risks since different economic entities have different cycles and structures. When one market behaves unfavourable, we may be still able to gain profit from another.

2.3. The Efficient Market Hypothesis (EMH)

The efficient market hypothesis is one of the most meaningful analyses in the finance field in the last twenty years. The efficient market hypothesis claims that stock price would adjust automatically when new information came into the market; this implies that the stock markets are integrated (Groppelli and Nikbakht, 2006). The degree of equity market efficiency has widely been used as an indication of integration. Under the assumption of information equivalence, expected risk and return will alter more consistently (Milne, 2004). No matter the conditions or recessions, every security should reflect its value accurately. Over time, benefits come from information asymmetric will no longer exist, and equity markets will finally become integrated (Groppelli and Nikbakht, 2006). While Bhattacharya and Constantinides (1989) hold an opposite view about the efficient market hypothesis; they believe that arbitrage opportunities are available as not every stock markets have the nature of these prerequisites.

3. Literature review

In the last decades, numerous pieces of literature estimated integration and co-movement among national stock markets. While earlier researches show that there is no support of long-run relationship was found. Chan et al. (1992) evaluate the relationships among equity markets in Singapore, Taiwan, Hong Kong, South Korea and the US during the 1980s period using daily and weekly data. By employ unit root and cointegration test, the result shows that there is no evidence of cointegration among the national share prices. This result reveals the success of universal diverseness among markets. Correspondingly, Ripley (1973) and Lessard (1976) also emphasise the advantages of invest in different stock markets to minimise portfolio risks as the research result reveal the low possibilities of stock market co-movement.

Some articles, in particular, analyse the cointegrating relationship of Asian stock market with inconsistent findings. The world’s first study regarding the common stochastic trends of Asian stock market is possibly conducted by Chung and Liu (1994). Chung and Liu use weekly data from 1 July 1985 to 18 May 1992, find two integrating relationships among six national indexes (including Taiwan, Singapore, Japan, Hong Kong, South Korea and the US). On the contrary, DeFusco et al. (1996) suggest that capital markets are segmented and there is no evidence of cointegration between the US and Asian markets (Thailand, Taiwan, South Korea, Malaysia and Philippines) using weekly data throughout January 1989 to May 1993. Similarly, Hee Ng (2002) build a system including five Association of Southeast Asian Nations (Indonesia, Philippines, Singapore, Indonesia and Thailand) examine monthly data from December 1988 to December 1997 and find non-existence of interdependency using Johansen (1988) multivariate test. Manning (2002) using both US dollar and local currencies dominated weekly data, as well as quarterly data during 1988 to 1999, illustrate the partial evidence of two common stochastic trends among Hong Kong, Singapore, Japan, South Korea, Indonesia, Phillippines, Thailand, and Malaysia.

The topic of stock market co-movement always been frequently mentioned or discussed by investors and researchers when particular financial events happen. For example, after the financial crisis in 1997 and 2008, people believed that the financial crisis would affect economics and stock market continuously and systematically, and the integration of financial can be captured by statistical methods to diversify their investment portfolios in the future. Jang and Sul (2002) divided the period of data into three stages, before, during and after the 1997 Asian financial crisis. By employ Granger causality model with a cointegration test, Jang and Sul (2002) examine whether the Asian stock markets (including Korea, Hong Kong, Singapore, Thailand, Japan, Indonesia and Taiwan) has co-movement. The results imply that no evidence of market cointegration was found between the seven Asian countries before the financial crisis. However, the linkage of Asian equity markets increased significantly from the starting time of the crisis in June 1997. Specifically, the co-movement among Hong Kong, Singapore, Thailand and Indonesia is remarkable. Further, eight months after the financial crisis, the connection between the seven national indexes present the most robust linkage occasionally. Our null hypothesis is that there is no cointegration between Hong Kong and Singapore stock market and the alternative hypothesis is stock market co-movement only happen when special events occur.

4. Data

The data used are adjusted closing prices for the period 1 January 1988 to 1 December 2019 consisted of 384 observations downloaded from Yahoo Finance. The indices used for stock market of Hong Kong and Singapore were the Hang Seng Index (HIS) and Straits Times Index (STI), separately. Hang Seng Index (HIS) is the most widely used indicator of Hong Kong stock market, tracks the top fifty firms listed on the Hong Kong Stock Exchange whereas Straits Times Index (STI) is the benchmark index for the Singapore stock market, which is made up of 30 constituents. Monthly data was selected since quarterly, and annual statistics always have spurious issues while daily data may also be influenced by the day of the week effect and its noise characterising problem (Roca, 1999).

Procedure: We transferred all the data into logarithmic forms before carrying out researches. 

5. Methodology

5.1. Unit Root Test

Before conducting a financial series analysis, it is vital to check the stationarity of data. Spurious regression may be generated due to non-stationary statistics, which indicate results may be biased. To address this problem, we use the Augmented Dickey-Fuller (ADF) test to check if the log-transformed adjusted closing price of Hang Seng Index and Straits Times Index (STI) have a unit root:

  $y_t=\mathit{ \mu }+y_{t-1}+u_t\mathit{ }\left(1\right)$

Where $y_t$ represents stock market index, $\mu$ is a constant parameter and $u_t$ is a disturbance term. Our null hypothesis is $y_t$ has a unit root and alternative hypothesis is there is $y_t$ does not have unit root (Hannan, 1983).

5.2. Cointegration Test

After completing the unit root test, the results suggest that both variables are non-stationary. We first differenced each variable and got significant statistics, which indicate the two variables are stationary at first differenced level. Consequently, we can move on to the cointegration test. In this paper, Engle and Granger Cointegration Test is employed to investigate the co-movement of Hong Kong and Singapore stock market:

$y_t=\alpha +\beta x_t+{\epsilon }_t\mathit{ }\left(2\right)$

where $y_t$ is Straits Times Index (STI), $\alpha$ is the constant term,$x_t$ is the Hang Seng Index (HSI) price and ${\epsilon }_t$ represents residual term. The null hypothesis is no cointegration between Hang Seng Index (HSI) and Straits Times Index (STI); alternative hypothesis is there are evidence of long-run relationship between them (Hannan, 1983).

6. Test results

6.1. Unit root test results

Table 1 demonstrates the unit root test results. It is shown that both the Hang Seng Index (HSI) and Straits Times Index (STI) share price is non-stationary in logarithmic form, but stationary in first differenced level. The results imply that the two indices are integrated of order one. Subsequently, we could perform a co-integration test for Hong Kong and Singapore stock market.

6.3. Co-integration test results

Table 2 illustrate the co-integration test results. It is seen that there is no co-integration between Hong Kong and Singapore national indices. Specifically, by comparing the t-statistics with critical values supplied by Engle and Yoo (1987), there is no evidence of stock price co-movement can be found. (t=-2.597288, p<-3.02).

7. Discussion on the empirical results

Three reasons may cause the absence of co-movement between Singapore and Hong Kong. Firstly, the insignificant results of the cointegration test imply that the possibilities of diversification of an international portfolio in stock markets are efficient for investors of Hong Kong and Singapore. These results are in line with the findings of Wong (2004) that also did not find a long-run relationship between Singapore and Hong Kong. However, Srinivasan, Kalaivani and Devakumar (2013) claim that there is cointegration among major Asia-Pacific economic entities (including Hong Kong, Singapore, China, South Korea, Taiwan, Japan and Indonesia).

Secondly, each Hong Kong and Singapore have closer neighbours, which are China and Southeast Asian Nations (ASEAN), respectively. The “China factor” in the Hong Kong stock market should not be underestimated. Su, Chong and Yan (2007) explore the possible share prices co-movement between Mainland China and Hong Kong indices before, and after the introduction of the Closer Economic Partnership Arrangement (CEPA) in 2003. They state that a massive number of A shares and H shares start to have a cointegrating relationship after the agreement was launched. Moreover, the establishment of the Shanghai-Hong Kong Stock Connect Program (SHSCP) in 2014 further deepened the influence of Mainland China on the Hong Kong stock market. This assumption is supported by the study from Xu (2017). 

Association of Southeast Asian Nations (ASEAN) was set up in 1967, initially consist of five countries, comprise as Singapore, Indonesia, Malaysia, Philippines and Thailand (ASEAN | ONE VISION ONE IDENTITY ONE COMMUNITY, 2019). According to the investing report by ASEAN (2019), the total GDP of its constitute reached 2.8 trillion dollars in 2017, which make them became the sixth-largest economy globally. Shabri Abd. Majid (2009) point that although ASEAN countries are cointegrated with each other before the 1997 Asian financial crisis, at the time when the financial crisis happen, ASEAN stock market co-movement reached a peak. 

8. Conclusion

This paper analyses the stock market co-movement between Hong Kong and Singapore. Using monthly data from 1988 to 2019 of the Hang Seng Index (HSI) and the Straits Times Index (STI), we conclude that there is no cointegrating relationship between the two indices. This result indicates that investing in Hong Kong is a decent choice for investors to diversify their portfolio who hold a substantial investment in Singapore, and vice versa. Future research can investigate more frequent data, such as daily or weekly. The other directions for future research are to analyse the spillover effect using volatility models.

9. Limitations

There are two main limitations presented in this paper. Firstly, we chose a long period as a sample without taking into account the impact of events that might have occurred during that period that would have affected the outcome. Secondly, the study can be improved by dividing the time into stages to draw more specific conclusions.

Reference

• ASEAN | ONE VISION ONE IDENTITY ONE COMMUNITY. (2019). ASEAN | ONE VISION ONE IDENTITY ONE COMMUNITY. [online] Available at: https://asean.org/ [Accessed 13 Dec. 2019].
• Bhattacharya, S. and Constantinides, G. (1989). Financial markets and incomplete information. Totowa, N.J.: Rowman & Littlefield.
• Chan, K., Gup, B. and Pan, M. (1992). An Empirical Analysis of Stock Prices in Major Asian Markets and the United States. The Financial Review, 27(2), pp.289-307.
• Chen, S. (2016). Interdependence of Property Prices and Building Vacancy Rates in Residential and Commercial Real Estate Markets: Hong Kong and Singapore. SSRN Electronic Journal.
• Chung, P., & Liu, D. (1994). Common stochastic trends in Pacific-Rim stock markets.
• Quarterly Review of Economics and Finance,34(5), 241−259.
• Groppelli, A. and Nikbakht, E. (2006). Finance. Hauppauge, N.Y.: Barron's.
• Hee Ng, T. (2002). Stock Market Linkages in South-East Asia. Asian Economic Journal, 16(4), pp.353-377.
• DeFusco, R. A., Geppert, J. M., & Tsetsekos, G. P. (1996). Long-run diversification potential in emerging stock markets.The Financial Review, 31(2), 343−363.
• Fabozzi, F. and Peterson, P. (2003). Financial management and analysis. Hoboken: Wiley.
• Hagin, R. (2004). Investment management. Hoboken, N.J.: Wiley.
• Hannan, E. (1983). Time series analysis. London: Chapman and Hall.
• Hee Ng, T. (2002). Stock Market Linkages in South-East Asia. Asian Economic Journal, 16(4), pp.353-377.
• IMD World Competitiveness ranking 2019. (2019). [ebook] IMD WORLD COMPETITIVENESS CENTER, p.1. Available at: https://www.imd.org/contentassets/6b85960f0d1b42a0a07ba59c49e828fb/one-year-change-vertical.pdf [Accessed 13 Dec. 2019].
• Investing in ASSEAN. (2019). [ebook] Association of Southeast Asian Nations, p.5. Available at: https://asean.org/storage/2019/10/Investing_in_ASEAN_2019_2020.pdf [Accessed 12 Dec. 2019].
• Jang, H. and Sul, W. (2002). The Asian financial crisis and the co-movement of Asian stock markets. Journal of Asian Economics, 13(1), pp.94-104.
• Lessard, D. (1976). World, Country, and Industry Relationships in Equity Returns: Implications for Risk Reduction through International Diversification. Financial Analysts Journal, 32(1), pp.32-38.
• Manning, N. (2002). Common trends and convergence? South East Asian equity markets 1988–1999.Journal of International Money and Finance,21(2), 183−202.
• Milne, F. (2004). Finance theory and asset pricing. Oxford: Oxford University Press.
• Ngiam, T. (2013). Developing Singapore as an international medical centre. Singapore Medical Journal, 54(11).
• Odier, P. and Solnik, B. (1993). Lessons for International Asset Allocation. Financial Analysts Journal, 49(2), pp.63-77.
• Ripley, D. (1973). Systematic Elements in the Linkage of National Stock Market Indices. The Review of Economics and Statistics, 55(3), p.356.
• Roca, E. (1999). Short-term and long-term price linkages between the equity markets of Australia and its major trading partners. Applied Financial Economics, 9(5), pp.501-511.
• Shabri Abd. Majid, M., Kameel Mydin Meera, A., Azmi Omar, M. and Abdul Aziz, H. (2009). Dynamic linkages among ASEAN‐5 emerging stock markets. International Journal of Emerging Markets, 4(2), pp.160-184.
• Srinivasan, P., Kalaivani, M. and Devakumar, C. (2013). Stock Market Linkages in Emerging Asia-Pacific Markets. SSRN Electronic Journal.
• Su, Q., Chong, T. and Yan, I. (2007). On the convergence of the Chinese and Hong Kong stock markets: a cointegration analysis of the A and H shares. Applied Financial Economics, 17(16), pp.1349-1357.
• Wong, W., Penm, J., Terrell, R. and Lim, K. (2004). The relationship between stock markets of major developedcountries and Asian emerging markets. Journal of Applied Mathematics and Decision Sciences, 8(4), pp.201-218.
• Xu, Y., Li, S., Xiong, X. and Ren, F. (2017). Intraday Volatility Spillover between the Shanghai and Hong Kong Stock Markets—Evidence from A+H Shares after the Launch of the Shanghai-Hong Kong Stock Connect. Journal of Management Science and Engineering, 2(4), pp.290-317.
• Young, A. (1992). A Tale of Two Cities: Factor Accumulation and Technical Change in Hong Kong and Singapore. NBER Macroeconomics Annual, 7, p.13.