australia's equity home bias - Wiley Online Library

Blackwell Melbourne, Australian AEPA © 0004-900X ORIGINAL XXX AUSTRALIA’S AUSTRALIAN Blackwell Publishing Economic Australia Publishing ARTICLES ECONOMIC EQUITY Asia Papers Ltd/University HOME PAPERS BIASof Adelaide and Flinders University of South Australia 2008

AUSTRALIA’S EQUITY HOME BIAS ANIL V. MISHRA University of Southern Queensland

This paper constructs the float adjusted measure of home bias and explores the determinants of Australia’s equity home bias by employing the International Monetary Fund’s high quality dataset (2001 to 2005) on cross border equity investment. On the empirical front, the paper conducts robustness tests by employing instrumental variables that are standard in the financial economics literature. The paper finds that the share of the number of firms listed in the domestic market and the share of internet users in the total population of the host country has a significant impact on equity home bias. Trade linkages are found to have a mixed impact on equity home bias. The paper also finds that the country’s market share of the world market capitalisation and transaction costs do not impact Australia’s equity home bias. Investors are found to exhibit low diversification motives.

Keywords JEL Classification coordinated s G11 portfolio G15 G30investment survey float instrumental variables. I. Introduction :

:

,

,

;

;

The traditional international capital asset pricing model (ICAPM) based on Sharpe (1964) and Lintner (1965) predicts that an investor should hold equities from a country as per that country’s share of world market capitalisation. However, empirical facts suggest that international portfolios are heavily biased towards domestic assets (French & Poterba, 1991; Cooper & Kaplanis, 1994; Tesar & Werner, 1995; Ahearne et al., 2004). This phenomenon is termed the ‘home bias’ and it can be defined as the situation where an investor holds far too high a share of their wealth in domestic securities compared with the optimal share predicted by the traditional theory of portfolio choice. For instance, the actual domestic equity holding of Australia in 2002 was 81.67 per cent, whereas the ICAPM benchmark percentage was 1.84. The empirical investigation into the home bias puzzle is important for several reasons. First, one of the major problems in the research on home bias has been the relatively poor quality of data on cross-border holdings. In the past, the cross-border holdings were estimated using accumulated capital flows and valuation adjustments (Tesar & Werner, 1995). Capital flows data were based on the balance of payments accounting; they record only the country of the foreign intermediary, which is not necessarily the country of the issuer. Warnock (2002) shows that capital flows data are ill suited to estimate bilateral holdings. This paper contributes to the existing literature by employing the International Monetary Fund’s (IMF’s) Coordinated Portfolio Investment Survey (CPIS) dataset on bilateral equity holdings for the years 2001 to 2005. CPIS reports data on foreign portfolio asset holdings (divided into equity, long term debt, and short term debt) by the residence of the issuer. In 1997, the IMF conducted the first CPIS in which 29 countries participated; the next survey was conducted in 2001 in which 69 countries participated and now CPIS is being conducted on an annual basis.

doi: 10.1111/j.1467-8454.2008.00329.x Correspondence: School of Accounting, Economics and Finance, University of Southern Queensland, Toowoomba Qld 4350. [email protected]. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

54

AUSTRALIAN ECONOMIC PAPERS

MARCH

Second, there are several papers investigating the home bias puzzle related to individual countries, viz. Japan (Kang & Stulz, 1997), Sweden (Dahlquist & Robertsson, 2001), Korea (Kim & Wei, 2002) and the United States (Ahearne et al., 2004; Dahlquist et al., 2003). Mishra and Daly (2006) and Mishra (2007) study Australia’s cross border portfolio investment using CPIS data. There is no study that exclusively focuses on Australia’s equity home bias. This is the first study that focuses exclusively on Australia’s equity home bias. Third, traditional studies on home bias assume that portfolio investors can hold a world market portfolio. However, Dahlquist et al. (2003) state that portfolio investors can only hold the float adjusted world market portfolio, i.e. a world portfolio of shares not held by insiders. This paper contributes to the existing literature on home bias, by constructing float adjusted measure of Australia’s home bias for the years 2001 to 2005. Fourth, this paper also empirically investigates the role of direct barriers, viz. transaction costs, information costs arising due to the share of internet users in the total population of the host country, the share of number of firms listed in the domestic market of the host country, the share of country’s stock market in world market capitalisation, trade patterns, and diversification motives – historical risk adjusted returns – and covariance on Australia’s equity home bias. Fifth, overall this paper fills in the gap by empirically investigating the phenomenon of home bias in the Australian context; which is critical in understanding international portfolio positions and capital flows. This paper provides answers to the following: How to construct float adjusted measure of home bias. Which factors are important in explaining Australia’s equity home bias? Does local market share of world capitalisation impact Australia’s equity home bias? Does the share of foreign equity listed in the domestic market impact equity home bias? Do trade linkages provide explanation for Australia’s equity home bias? Do share trading costs associated with destination countries’ stock exchanges affect Australia’s equity home bias? Do investors’ diversification motives primarily influence Australia’s equity home bias? This paper constructs float adjusted measure of Australia’s equity home bias. Results indicate that share of internet users in the total population and share of foreign equity listed in the domestic market has significant impact on Australia’s equity home bias. Trade links provide mixed results for Australia’s equity home bias. Local market share of the world market capitalisation and transaction costs do not impact Australia’s equity home bias. Investors’ exhibit low diversification motives. This paper is structured as follows: Section II provides literature review of the home bias puzzle. Section III describes float adjusted measure of home bias. Section IV describes the sources of home bias. Section V discusses theoretical framework, empirical specification and instrumental variables. Section VI describes the empirical results and, finally, Section VII provides a conclusion.

II.

L i t e r at u r e R e v i e w 1

Black (1974) and Stulz (1981) develop a two-country capital market equilibrium model where there are barriers to cross border investment and these barriers can be considered as tax on net foreign investment. This tax represents various kinds of barriers to international investment, such as direct controls on the import or export of capital, possibility of expropriation of foreign holdings, reserve requirements on bank deposits and other assets held by foreigners, restrictions on the fraction of business that is owned by foreigners. It may also include barriers due to information asymmetries, i.e. unfamiliarity of residents of one country with 1

Lewis (1999) and Karolyi and Stulz (2003) provide excellent reviews of the home bias literature.

© 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008

AUSTRALIA’S EQUITY HOME BIAS

55

the stock markets of other countries. Merton (1987) develops a model where investors hold stocks that they know. In this model, investors believe that the risk of stocks they do not know is extremely high. Accordingly, the investors may overweight domestic stocks. Cooper and Lessard (1981) developed an international capital market equilibrium model which allows for differential taxes on foreign investment depending on the country of investment and the origin of investor. Cooper and Lessard (1981) create unique solutions for taxes under extreme assumptions that taxes depend on the country of investment, or on the origin of investor. Cooper and Kaplanis (1994) find that hedging against inflation risk cannot explain the home bias. Several research papers have considered the effect of indirect barriers, such as information asymmetries, on equity investment and home bias. French and Poterba (1991), for instance, find that information asymmetry can generate the same observed portfolio patterns when investors expect the domestic returns to be several hundred basis points higher than the returns in foreign markets. Gehrig (1993) uses a noisy rational expectations model to investigate the effect of asymmetric information between domestic and foreign investors. Investors observe noisy signals with different degrees of precision. Domestic investors receive signals of future returns that are more precise. Investors remain incompletely informed, even in equilibrium. Domestic bias arises from better investor information about domestic stocks. Thus, on average, foreign investments appear to be more risky. Hasan and Simaan (2000) derive the premium that an investor is willing to pay to buy the full information of the mean return vector and show that rational investors prefer home country dominated portfolios over diversified portfolios if the variability of estimation errors far exceeds the variability of the mean return vector. Coval and Moskowitz (1999, 2001) show that the weight of a US stock in US mutual funds is negatively related to the distance between the location of the fund and the location of the headquarters of the firm. The mutual fund managers do better with their holdings of stocks of firms located more closely to where the mutual fund is located. Portes et al. (2001) find that information asymmetries are responsible for the strong negative relationship between asset trade and distance. They investigate the roles of explicit information variables, including distance, in explaining separately cross-border trade in corporate equities, corporate bonds, and government bonds. Sarkissian and Schill (2004) find that geographic, economic, cultural, and industrial proximity plays a dominant role in the selection of overseas listing stock exchange. Their findings imply that proximity constraints that lead to home bias in investment portfolio decisions are similar to those which influence financing decisions. Amadi (2004) states that there has been a distinct reduction in equity home bias in recent years. The rise of the internet and mutual fund investment has affected changes in foreign diversification, supporting information asymmetries explanation. Market size, which intuitively might be expected to affect foreign diversification, is insignificant. Portes and Rey (2005) explore a new panel data set on bilateral gross cross-border equity flows between 14 countries, for a period from 1989 to 1996. They show that gross transaction flows depend on market size in source and destination country, as well as trading costs, in which both information and the transaction technology play a role. In their model, distance proxies some information costs, and other variables explicitly represent information transmission, an information asymmetry between domestic and foreign investors, and the efficiency of transactions. They find that the geography of information is the main determinant of the pattern of international transactions, while there is weak support for diversification motive, in their data, once they control for the information friction. Chan et al. (2005) examine how mutual funds from 26 developed and developing countries allocate their investment between domestic and foreign equity markets and what factors determine their asset allocation worldwide. They state that the stock market development and familiarity variables have significant, but asymmetric, effects on the domestic bias and foreign bias and that economic development, capital controls, and withholding tax variables have significant effects only on the foreign bias. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

56


MARCH

For Japan, Kang and Stulz (1997) find that foreign investors concentrate on equity investments in firms that are large; firms that export more and firms with good accounting performance. For Sweden, Dahlquist and Robertsson (2001) find that non-resident investors are mostly institutional investors and that the holdings of stocks by non-resident investors exhibit biases that are also typical of resident institutional investors. Their findings are consistent with Kang and Stulz (1997). Grinblatt and Keloharju (2001) state that investors are more likely to hold, buy, and sell the stocks of Finnish firms that are located close to the investor, that communicate in the investor’s native tongue, and that have chief executives of the same cultural background. For Korea, Choe et al. (2001) find that foreign investors buy at higher prices than resident investors and sell at lower prices. Kim and Wei (2002) find that a significant information asymmetry exists between the resident foreign investors and non-resident foreign investors. They base their finding by testing the hypothesis that non-resident foreign investors may hoard more than resident foreign investors, for example, Korean subsidiaries and branches of foreign institutions, where the latter group have more timely information about the country they live in. Hau (2001) finds that proprietary trades on the German stock market do better when they are geographically closer to Frankfurt. For US, Ahearne et al. (2004) test the home bias puzzle by employing the data on US holdings of foreign equities. They find that information cost is a major determinant of a country’s weight in US investor’s portfolio. For Australia, Mishra and Daly (2006) state that the major determinants of Australia’s geographical allocation of portfolio investment indicate a broad correspondence between stock market capitalisation of destination countries and the allocation of Australian financial investments, but with some deviations from that baseline, where the deviations are correlated with Australian trade patterns. Mishra (2007) examines the bilateral, source and host factors driving portfolio equity investment across a set of countries using CPIS data on international equity holdings at the end of 1997, 2001 and 2002. He states that the bilateral equity investment is strongly correlated with the underlying patterns of trade in goods and services. The information asymmetries and cultural-institutional proximity are important for bilateral equity investment. The size of domestic stock market is the key correlate of aggregate foreign portfolio equity asset and liability holdings. The scale of aggregate foreign equity asset holdings is larger for richer countries. Dahlquist et al. (2003) state that there is a close relation between corporate governance and the portfolios held by investors. They show that US investors underweight those foreign countries in their portfolios which have closely held firms. They construct an estimate of the world float portfolio. They also analyse Swedish firm level data on foreign ownership and closely held shares, and show that the weight of a Swedish firm in the portfolio of foreign investors is inversely related to the fraction of firm held by controlling shareholders. Kho et al. (2006) find that the home bias of US investors decreased the most towards countries in which the ownership by corporate insiders is low, and countries in which ownership by corporate insiders fell. Using firm-level data for Korea, they find that portfolio equity investment by foreign investors in Korean firms is inversely related to insider ownership and that the firms that attract the most foreign portfolio equity investment are large firms with dispersed ownership.

III.

F l o at A d j u s t e d H o m e B i a s

Suppose the source country is i and the host country is j. Share of i’s equity in country j ( I i j ) is the ratio of i’s holdings of country j equities to country i’s total equity portfolio. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008


Iij =

Country i’s holdings of country j equities Country i’s total equity portfolio

Country i’s total Investment by country i’s Investment by country i’s = + equity protfolio residents in home equities residents in foreign equities

57

(1) (2)

In this paper, country i is the source country, Australia. In other words, Australia’s total equity portfolio is investment by resident Australians in home equities, plus investment by Australians in foreign countries. Investment by country i’s Country i’s maket Country i’s equities held = − residents in home equities capitalisation by foreign investors

(3)

In other words, investment by resident Australians in home equities is the difference between Australia’s market capitalisation and Australia’s equities held by other countries. The market capitalisation value is determined from Federation Internationale des Bourses de Valeurs (FIBV) database of World Stock Exchanges. The equity data is from IMF’s CPIS dataset on cross border portfolio equity investment for the years 2001 to 2005. Appendix A lists countries with cross border equity investment (CPIS data) in Australia and Australia’s equity investment abroad, from 2001 to 2005. Sharpe’s (1964) and Lintner’s (1965) models are based on perfect markets. Their models assume that investment and consumption opportunity sets do not differ across countries and that investors are the same across countries with respect to risk aversion and information. There are no barriers to international investment, no restrictions on short sales, no taxes, no information asymmetries and no tariffs. The traditional ICAPM model suggests that to maximise risk adjusted returns, investors should hold equities from countries around the world in proportion to their market capitalisation. It follows that share of country i ’s equities invested in country j ( I *) j , is the ratio of market capitalisation of country j in the world market capitalisation. I *j =

MC j MC world

(4)

where MCj is the market capitalisation of country j and MCworld is the world market capitalisation. This ratio is the benchmark of portfolio holdings to which the actual portfolio share is compared. Ahearne et al. (2004) employ the traditional approach to measure home bias in United States. The equity home bias is the deviation from the ICAPM benchmark, defined as one minus the ratio of foreign equities in the US and world portfolios. HBij = 1 −

Iij I *j

(5)

The traditional theory of home bias calculates the world market portfolio assuming that all shares issued by a corporation could potentially be held by foreign investors. Dahlquist et al. (2003) state that in countries with poor investor protection, firms tend to be controlled by large shareholders so that foreigners can hold only a small portion of issued shares that are freely traded or floated. Firms outside the United States are typically controlled by large resident shareholders (La Porta et al., 1999). These large resident share holders are the controlling share holders, who only sell their shares as a control bloc for a price significantly above the open market share trade prices. Shares held by the controlling share holders are also known as © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

58


MARCH

closely held shares. The controlling shareholder would not sell their shares without being paid a premium to reflect the benefits they derive from control. The controlling shareholders may be officers, directors, and their immediate families, and the shares may constitute shares held in trusts, shares held by pension benefit plans, and shares held by individuals who hold five per cent or more of the outstanding shares. The Japanese closely held shares represent the holdings of the ten largest shareholders. Suppose the controlling share holders insider ownership is α. Portfolio investors can only hold shares in a firm, not held by the controlling shareholders. Portfolio investors (or noninsiders) can hold (1 − α) of the firm. Foreign investors can only hold a fraction a of the share held by non-insiders. Foreign investors hold a(1 − α) of the firm if they have no home bias. Dahlquist et al. (2003) state that portfolio investors cannot hold the world market portfolio, but can only hold the world market portfolio of shares not held by insiders; which is also known as the float adjusted world market portfolio. If all investors hold the float adjusted world market portfolio, then as insider holdings fall, foreign investors can buy a fraction of shares sold by insiders equal to the weight of the country in the float adjusted world market portfolio. But if foreign investors do not hold the float adjusted world market portfolio, then there is no necessary relation between a change in insider ownership and a change in shares held by foreign investors because all the shares sold by insiders could be bought by local investors. This paper calculates the float adjusted portfolio for countries and also float adjusted world market portfolio from DataStream’s Worldscope database. The float adjusted market capitalisation for a country is the sum of the values of free float market capitalisation for all the firms in that country. Free float market capitalisation is free float number of shares multiplied by the latest available share price, in millions of currency units. Free float number of shares is the percentage of total shares in issue available to ordinary investors, i.e. the total number of shares less the closely held shares. Appendix B provides annual data on the number of firms with free float market value. MVFF j =

∑m MVFF jm

(6)

where MVFFj is the float adjusted market capitalisation for country j, MVFFjm is the free float market capitalisation of firm m in country j and ∑ m MVFF jm is the sum of free float market capitalisation of all firms in country j. MVFFworld =

∑j MVFF j

(7)

where ∑ j MVFF j is the sum of free float market capitalisation for the countries in the world. Appendix B lists the countries whose free float market capitalisation data is used for calculating the world float adjusted market capitalisation. Upon incorporating free float measures, equation (4) becomes I *FF, j =

MVFF j MVFFworld

(8)

j I FF ,i I *FF, j

(9)

Finally, the free float home bias measure is HBFF, ij = 1 −

j where HBFF,ij is the float adjusted measure of home bias, I FF , i is the float adjusted measure of country i’s equity holdings in country j and I *FF, j is float adjusted world market portfolio of country j.


2008


59

Table I

Australia’s portfolio equity investment (2004)

Country

(1) Actual share in Australia’s equity portfolio

(2) Benchmark share in world float market capitalisation

(3) Actual over Benchmark

(4) HBFF,ij

0.026 0.354 0.001 0.061 0.065 0.525 0.288 1.661 0.184 0.005 0.023 1.227 0.058 0.054 0.122 0.035 0.193 0.139 0.025 1.746 11.076

0.147 2.462 0.035 0.373 0.631 2.444 2.612 11.719 0.817 0.151 0.322 1.834 0.095 0.198 0.765 0.863 2.721 1.122 0.254 4.573 32.836

0.176 0.143 0.030 0.164 0.103 0.215 0.110 0.141 0.226 0.033 0.073 0.669 0.609 0.272 0.160 0.041 0.071 0.123 0.098 0.381 0.337

0.824 0.857 0.970 0.836 0.897 0.785 0.890 0.859 0.774 0.967 0.927 0.331 0.391 0.728 0.840 0.959 0.929 0.877 0.902 0.619 0.663

Austria Canada Czech Republic Denmark Finland Germany Hong Kong Japan Korea Luxembourg Malaysia Netherlands New Zealand Norway Singapore South Africa Spain Sweden Thailand UK US

Note: HBFF,ij: Float adjusted home bias measure. Source: Foreign equity investments from the IMF’s CPIS, market capitalisations from FIBV. (Author’s own calculation.)

Home bias is equivalent to normalising source country holdings in the host country by the country’s float market capitalisation and then dividing by the share of host country holdings in the worldwide float market capitalisation. The empirical analysis in this paper employs the float adjusted home bias measure for the years 2001 to 2005. Table I presents Australia’s float adjusted home bias measure as of December 2004. Column (1) of the table presents Australian investors’ actual portfolio share as of December 2004. The actual portfolio share is the foreign equity holdings of Australia in other countries relative to Australia’s total holdings of foreign and domestic equities. Column (1) indicates that Australia’s actual per cent portfolio share is the highest in US (11.07) followed by UK (1.74), Japan (1.66), Netherlands (1.22) and then, the remaining countries of the world. Column (2) of the table presents the theoretical portfolio shares, i.e. share of country’s float market capitalisation in the world float market capitalisation. It shows the share of Australia’s equity holdings by country under the assumption that investors choose portfolios based on the standard portfolio theory. Column (3) compares the actual share of domestic equities held by Australians in other countries with the benchmark share in the world portfolio as per ICAPM model. This comparison gives an indication of the degree to which Australian investors’ underweight different foreign countries. Column (3) clearly indicates that there is a significant amount of variation in values across countries and Australian holdings are less than those predicted by ICAPM. The ratio is 0.66 for Netherlands, indicating that Australian investors’ holding of stocks from Netherlands at end-2004 was 66 per cent of what traditional portfolio theory would have predicted. The degree of underweighting is more severe against countries like Czech Republic (0.03), where Australian investor holds three per cent of the shares predicted by traditional ICAPM levels. Column (4) indicates the measure of home bias as per © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

60


MARCH

equation (9). A greater value of home bias measure corresponds to a lower weight in Australia relative to world portfolios and, thus, a higher degree of bias.

IV.

Sources of Home Bias

Some of the possible sources of home bias in the Australian investor’s equity holdings may be due to explicit costs, familiarity and diversification motives. These sources of home bias are discussed below.

a) Explicit costs Black (1974), Stulz (1981), Cooper and Kaplanis (1986) and Ahearne et al. (2004) state that investors face explicit costs related to equity investment, viz. transaction fees, taxes, commissions and the costs of gathering information.

i) Transaction costs (TRAN) Home bias can arise due to high transaction costs associated with trading foreign equities. The transaction cost data is derived from Elkins-McSherry Co (www.elkins-mcsherry.com). Elkins-McSherry Co receives trade data on all global trades by institutional traders and computes measures of trading costs. The data consists of average trading costs as a percentage of trade value for active managers in a universe of 42 countries. The data are quarterly, from the last quarter of 1995 through to the fourth quarter of 2006. In 1998, the institutional traders in the data represented 136 firms, of which 105 were pension funds, 27 were investment managers, and 4 were brokers. These institutions accounted for 28 billion shares in 632 547 trades, using 700 global managers and 1000 brokers (Domowitz et al., 2001). The transaction cost comprises of three cost components, viz. commissions, fees and market impact costs. This paper takes into account the total cost comprising all three cost components for the end quarter of years 2001 to 2005. Investors would underweight high transaction cost countries in their portfolios and, accordingly, this variable is expected to have a positive impact on the measure of home bias.

b) Familiarity Mishra (2007), Mishra and Daly (2006), Portes and Rey (2005), Sarkissian and Schill (2004), Aviat and Coeurdacier (2007), Coval and Moskowitz (1999, 2001), Portes et al. (2001), Grinblatt and Keloharju (2001) and Huberman (2001) find that familiarity plays a dominant role in investor’s preference for equity investment. The following section discusses familiarity measures of home bias: internet, foreign listing, trade and size.

i) Internet (INT) Internet plays a major role in supplying financial information to investors through its financial websites. Investors have access to vast amounts of information on equity investment via internet, which may easily influence their investment behaviour. Internet may proxy for information asymmetries as it eases access to financial information, which investors may use for their portfolio diversification. Internet is the share of internet users in a country’s population. Amadi (2004) uses this measure and finds that the rise of the internet has affected the changes in foreign diversification. The data on internet users is from United Nations Statistics Division. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008


61

ii) Foreign listing (FL) Foreign listing is the share of the number of foreign firms listed in total number of firms listed in domestic market. Baker et al. (2002) state that firms that list publicly in the United States are more visible to US investors. Cross listing makes information more readily available and the cost of gathering information on foreign firms is also reduced. Local investors can readily access foreign equity of the listed foreign firms in their domestic markets. Ahearne et al. (2004) state that foreign countries whose firms do not alleviate information costs by opting into the US regulatory environment are more severely underweighted in US equity portfolios. Amadi (2004) uses this foreign listing variable and finds that foreign listing has an insignificant effect on overall foreign diversification of countries. The data on the number of foreign firms and total number of firms in domestic markets is from International Federation of Stock Exchanges.

iii) Trade (TRAD) Mishra (2007), and Lane and Milesi-Ferreti (2004) state that bilateral equity investment is strongly correlated with the underlying patterns of trade in goods and services. Trade is the average of imports and exports normalised by the destination country’s GDP. This measure is in accordance with Ahearne et al. (2004). Australian investors are better able to attain accounting and regulatory information on foreign markets through trade. Consequently, investors may be inclined to hold the stocks of foreign companies with whose products they are most familiar. This variable is expected to have a negative impact on the measure of home bias. The data on imports and exports is from IMF’s Direction of Trade Statistics and GDP data is from World Bank’s World Development Indicators.

iv) Size (SZ) SZ is country’s market share of the world market capitalisation. This variable tests the assumption of the traditional theory of ICAPM that investors should diversify according to their country’s share of world market capitalisation. This measure is in accordance with Ahearne et al. (2004) and Amadi (2004). The stock market data is from International Federation of Stock Exchanges.

c) Diversification Bohn and Tesar (1996) state that investors are momentum traders or return chasers, who base their equity investment decisions on stock markets past performance. They state that investors tend to move into markets where returns are expected to be high and retreat from markets when predicted returns are low. Brennan and Cao (1997) developed a model of international equity portfolio investment flows based on differences in informational endowments between foreign and domestic investors. They show that when domestic investors possess a cumulative information advantage over foreign investors about their domestic market, investors tend to purchase foreign assets in periods when the return on foreign assets is high and to sell when the return is low. This paper employs two measures, i.e. covariance and reward to risk ratio, to investigate the diversification motives of Australian investors.

i) Covariance (COV) The financial economics literature suggests that the greater the co-movements between financial assets of two countries, the lower the benefit of diversification. When the correlation between source country and host country is small, source country investors enjoy a larger diversification gain from investing in the host country; they have greater desire to increase their equity holdings © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

62


MARCH

in the host country. Therefore, the degree of home bias of source country for host country will be smaller. The covariance between source and host country is computed using return data from DataStream’s Morgan Stanley Capital International (MSCI). The return data is calculated from MSCI monthly stock market indices for months ranging from January 1995 to December 2005.

ii) Reward to risk ratio (RWRSK) Reward to risk ratio is the ratio of mean monthly return to standard deviation. This measure is in accordance with Ahearne et al. (2004). Investors might tend to underweight those countries in their portfolios, whose stock markets have performed poorly, based on their information of past stock returns. This variable is expected to have negative impact on the measure of home bias. The return data is calculated from Datastream’s Morgan Stanley Capital International (MSCI) monthly stock market indices for months ranging from January 1995 to December 2005.

V.

T h e o r e t i c a l F r a m e w o r k , E m p i r i c a l S p e c i fi c at i o n a n d I n s t r u m e n ta l V a r i a b l e s

a) Theoretical framework This paper is based on Cooper and Kaplanis’ (1986) theoretical framework. They derive efficient portfolios in a world where there are barriers to cross border investment, which depend both on the domicile of the investor and their country of investment. The i investor’s optimisation problem is to maximise expected returns net of costs: Max ( xi′R − xi′ci )

(10)

subject to xi′Vxi = v xi′I = 1 where xi is a column vector, xin is the nth element of xi and is the proportion of individual i’s total wealth invested in risky securities of country n, R is a column vector of pre-tax expected returns, ci is a column vector, the nth element of which is cin, cin is the deadweight cost to investor i of holding securities in country n, v is a constant, V is the variance/covariance matrix of the gross (pre-cost, pre-tax) returns of the risky securities, I is a unity column vector. The Lagrangean of above maximisation problem is

( 2 ) (x′Vx − v) − k (x′I − 1)

L = ( xi′R − xi′ci ) − h

i

i

i

i

(11)

where h and ki are Lagrange multipliers. Equating the derivative of the objective function with respect to xi equal to zero, R − ci − hVxi − ki I = 0

(12)

Therefore the optimal portfolio for investor i is −1 ⎞ xi = ⎛⎝V ( R − ci − ki I ) h⎠ −1 −1 ki = [ I ′V R − I ′V ci − h]

I ′V −1 I


(13) (14)

2008


63

Investors hold the world market portfolio, the minimum variance zero beta portfolio and a fund with minimum variance for a specified level of deadweight cost. The world capital market equilibrium is the aggregation of the individual portfolio holdings. The clearing condition for the model is

∑Wi xi = M

(15)

where Wi is the proportion of world wealth owned by country i, M is a column vector, the ith element of which is Mi, Mi is the proportion of world market capitalisation in country i’s market. Substituting equation (14) in equation (15), R−

∑Wi ci − ∑Wi ki I = hVM

(16)

Subtracting equation (16) from equation (13), hV ( xi − M ) =

(∑W c

i i

) (∑W k

− ci + I

i i

− ki

)

(17)

∑Wi ki − ki = z ′(ci − ∑Wi ci )

(18)

−1 z =V I

(19)

( I ′V −1 I )

where z is the global minimum variance portfolio. Substituting (18) in (17) gives hV ( xi − M ) =

(∑W c

i i

)

(∑W c

− ci + z ′

i i

)

− ci I

(20)

If dead weight costs are zero, then each investor holds the world market portfolio, since the right hand side of equation (20) equals zero. Consider now the case when the covariance matrix V is diagonal with all variances equal to s2 and the deadweight cost of any country/investor pair is equal to c, except for domestic investment where it is equal to zero. The portfolio holdings of investor i in country n are ⎛ ⎞ xin = M n − ⎜Wnc 2 ⎟ , i ≠ n ⎝ hs ⎠ ⎛ ⎞ xin = M n − ⎜Wnc 2 ⎟ + ⎛⎝ c 2 ⎞⎠ , i = n hs ⎠ hs ⎝

(21)

Equation (21) indicates that the larger the marginal deadweight cost c, the greater the deviation of portfolio holdings from the world market portfolio. This deviation is negative for foreign investment and positive for domestic investment. Investors put a greater weight on domestic securities and less weight on foreign securities. In the case of non-uniform deadweight costs, equation (20) can be expressed as, pi = qi − z′qi I pi = hV(xi − M) qi =

∑ M j c j − ci


(22)

64


MARCH

The nth equation of the system given by (22) can be expressed as, pin = −cin + bn + ai − d, i ≠ n

(23)

pin = bn + ai − d, i = n

(24)

ai = z′ci bn =

∑ M j c jn

d = z ′∑ M i ci where ai is the weighted average marginal deadweight cost for investor i, bn is the weighted marginal deadweight cost for investors investing in country n and d is the world weighted average marginal deadweight cost. In case, where the covariance matrix is diagonal with all variances equal to s2, equations (23) and (24) can be expressed as, hs2(xin − Mn) = −cin + bn + ai − d, i ≠ n

(25)

hs2(xin − Mn) = bn + ai − d, i = n

(26)

In the more general case with non-zero and non-uniform costs, if the actual cost to investor i of investing in country n (cin) is high relative to investor i’s average cost to investing (ai) or relative to all investors costs to investing in country n (bin), then the right hand side of equation (25) is likely to be negative and investor i will underweight country n in their portfolio. The higher the costs in a particular foreign market, the more severely underweighted that country will be in an investor’s portfolio. Equation (25) measures the extent to which the holdings of a source country investor in foreign markets deviate from those of the world market portfolio. Since there are no barriers for investor i when investing domestically, the right – hand side of equation (26) will be positive, and consequently the investor would be overweight relative to the market in domestic securities. The portfolio choice depends on the relative size of these costs to the weighted average marginal deadweight cost of the investor, or of the country of investment. One implication is that investors are likely to underweight countries for which it is costly to gather and analyse firm-level information.

b) Empirical specification The paper regresses the measure of home bias (discussed in Section III) on a vector of explanatory variables that include size, foreign listing, internet, trade, transaction cost, reward to risk ratio and covariance (discussed in Section IV). HBFF,ij = α0 + α1(SZ) + α2(FL) + α3(INT) + α4(TRAD) + α5(TRAN) + α6(RWRSK) + α7(COV) + χj (27) where HBFF,ij: Float adjusted measure of home bias, SZ: share of a country’s stock market in world market capitalisation, FL: share of foreign firms listed in the domestic market, INT: share of internet users in the total population, TRAD: Trade is the average of imports and exports normalised by the destination country’s GDP, TRAN: Transaction cost associated with share trading in destination country, RWRSK: Reward to risk is the ratio of destination country’s mean monthly return to standard deviation, COV: Covariance of monthly returns between source country and destination country, χj: random error term. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008


65

c) Instrumental variables This paper uses CPIS data on cross border equity holdings to calculate float equity home bias. The CPIS’s equity investment data is based on the concept of capital stock. Endogeneity should be less of a problem for capital stocks than flows. However, in order to address possible endogeneity problems, measurement errors and omitted variable bias, this paper adopts two stage least squares techniques, by employing instrumental variables that are standard in the literature. Return variables and transaction cost variables are endogenous and there is a possibility that they may correlate with the error terms. In accordance with Li et al. (2004), this paper employs market size variables (logarithm of the gross domestic product of host country and logarithm of the number of publicly listed companies in the host country) as instruments to return variables. This paper considers the index of religion from Alesina et al. (2003) as an instrument. Their religious fractionalisation variable, ‘religion’ is based on the data from Encyclopaedia Britannica (2001) and covers 294 different religions in 215 countries. ‘Religion’ identifies the percentage of the population of each country that belonged to the three most widely – spread religions (Roman Catholic, Protestant and Muslim) in the world in 1980. For countries of recent formation, the data is available for 1990 to 1995. The numbers are in per cent from zero to one hundred. Mishra and Daly (2007) use the ‘religion’ index as an instrument to determine the effect of quality of institutions on outward foreign direct investment. Further, this paper uses the index of ethnolinguistic fractionalisation from Mauro (1995) as an instrument. Mauro (1995) constructs the ethnolinguistic fractionalisation from raw data that come from Atlas Narodov Mira (Department of Geodesy and Cartography of the State Geological Committee of the USSR 1964). The raw data criteria for characterising groups as ethnically separate related mainly to historical linguistic origin, and not to any economic or political variables. Ethnolinguistic fractionalisation measures the probability that two persons drawn at random from a country’s population will not belong to the same linguistic group. Mauro (1995) uses this variable as an instrument to determine the extent to which government institutions affect economic growth. The extent to which countries are fractionalised along ethnolinguistic lines is exogenous and unrelated to economic variables. This paper also employs data on the quality of domestic institutions as instrumental variables because quality of institutions is largely exogenous to bilateral capital stocks. The quality of institutions data is from Kaufmann et al. (2006) indicators which describe various aspects of the governance structures of a broad cross-section of countries. The indicators have been constructed on the basis of information gathered through a wide variety of cross-country surveys, as well as polls of experts. The indicators cover 213 countries and territories for 1996, 1998 and 2000, and annually for 2002 to 2005. In 2005, they estimate the indicators on 276 individual variables measuring different dimension of governance, by employing 31 different datasets from 24 different sources. They construct six indicators each representing a different dimension of governance, viz. voice and accountability, political instability, government effectiveness, regulatory burden, rule of law and control of corruption. The variables are standardised to have a mean of zero and a standard deviation of one. The larger the values, the better they indicate their institutional quality. This paper uses two Kaufmann et al. (2006) indicators as instruments: rule of law and regulatory quality. Rule of Law (RL) index measures concepts related to the enforceability of government and private contracts, fairness of judicial process, speediness of judicial process, violent and organised crimes, trust in legal system, patent and copyright protection, etc. Regulatory Quality (RQ) index consists of indicators related to the regulations of exports, imports, business ownerships, equities ownerships, banking, foreign investment, price controls, tariffs, unfair competitive practices, etc. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

66 Table II

SZ FL INT TRAD TRAN RWRSK COV


MARCH

Correlation matrix (2001 to 2005) HBFF,ij

SZ

FL

INT

TRAD

TRAN

RWRSK

−0.21 −0.07 −0.27 −0.19 0.20 −0.10 0.01

−0.03 0.24 −0.16 −0.31 0.21 −0.13

0.17 0.03 −0.24 0.28 −0.62

0.07 −0.41 0.41 −0.25

0.20 −0.35 0.19

−0.15 0.54

−0.23

Notes: *, ** and *** indicate significance at the 1%, 5% and 10% levels, respectively. White corrected tstatistics in parenthesis. HBFF,ij: float adjusted home bias measure, SZ: share of a country’s stock market in world market capitalisation, FL: share of foreign firms listed in the domestic market, INT: share of internet users in the total population, TRAD: trade is the average of imports and exports normalised by the destination country’s GDP, TRAN: transaction cost associated with share trading in destination country, RWRSK: reward to risk is the ratio of destination country’s mean monthly return to standard deviation, COV: covariance of monthly returns between source country and destination country.

VI.

E m p i r i c a l R e s u lt s

The empirical results are based on panel regression and robustness tests using instrumental variables that are standard in the financial economics literature for the years 2001 to 2005. Table II presents the correlation matrix for the variables used in the paper. The measure of home bias is negatively related to the share of a country’s stock market in world market capitalisation, share of foreign firms listed in the domestic market, share of internet users in the total population, trade, reward to risk ratio; and positively related to transaction cost and covariance. The covariance variable (COV) is highly correlated (−0.62) with share of foreign firms listed in the domestic market variable (FL). In addition, the covariance variable (COV) is highly correlated (0.54) with the transaction cost variable (TRAN). Overall, the correlation matrix does not indicate serious correlation among the variables. Table III indicates the panel regression results of Australia’s home bias measure by regressing home bias variable against the independent variables, i.e. size, foreign listing, internet, trade, reward to risk ratio and covariance. Table III also indicates two stage least squares panel regression results as robustness tests. In column (1), size variable (SZ) enters significantly at the five per cent level with a negative sign. This is contrary to theory that home bias should increase as investors’ local market’s share of world market capitalisation increases. Internet variable (INT) is negative and significant at five per cent, implying that investors are able to invest in foreign equities based on the investment information gathered from the internet. Trade variable (TRAD) is negative and significant at five per cent, implying that Australian investors may be inclined to hold securities of close trading partners for a variety of reasons, including familiarity, hedging or availability of information. Covariance variable (COV) is positive and significant at ten per cent. Covariance variable is used to test the diversification motive. If transactions occur because of diversification motive, the covariance variable should be significant because the greater the co-movement between financial assets of two countries, the lower the benefit of diversification. Covariance variable, though statistically significant, is economically insignificant suggesting low diversification motives of investors. The overall adjusted R2 is 0.14. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008 Table III


67

Australia’s home bias (2001 to 2005) (1)

SZ FL INT TRAD

(2)

(3)

(4)

(5)

(6)

−0.26** (−2.33)

0.01 (0.03)

−0.15** (−2.00) −0.01** (−2.15)

−0.60* (−3.07) −0.03** (−2.25)

−0.29* (−2.77) −0.45*** (−1.66) −0.11*** (−1.70) −0.01*** (−1.74) 0.05 (0.39)

−0.26 (0.60) −0.91*** (−1.75) −0.61** (−2.12) −0.00 (−0.48) 1.85** (2.27)

−0.26** (−2.54) −0.43*** (−1.65)

−0.57 (−0.90) −0.67*** (−1.69)

−0.01** (−2.26)

−0.03** (−2.01)

0.00 (1.64) 0.80* (14.18)

0.00 (0.26) 0.89* (4.25) REL DL GDP RL RQ

RWRSK COV TRAN Constant

0.00*** (1.80) 0.84* (20.10)

Instrumental Variables

Adjusted R2 Observation

0.14 108

−0.00 (−0.06) 1.09* (7.43) REL DL GDP RL ELF 108

0.91* (34.29)

0.24 108

1.01* (11.56) REL DL GDP RL ELF 108

0.25 108

108

Notes: *,** and *** indicate significance at the 1%, 5% and 10% levels, respectively. White corrected tstatistics in parenthesis. HBFF,ij: float adjusted home bias measure, SZ: share of a country’s stock market in world market capitalisation, FL: share of foreign firms listed in the domestic market, INT: share of internet users in the total population, TRAD: trade is the average of imports and exports normalised by the destination country’s GDP, TRAN: transaction cost associated with share trading in destination country, RWRSK: reward to risk is the ratio of destination country’s mean monthly return to standard deviation, COV: covariance of monthly returns between source country and destination country, REL: index of religion from Alesina et al. (2002), DL: logarithm of the number of publicly listed domestic companies in the country, GDP: logarithm of gross domestic product, ELF: ethno linguistic fractionalisation index, RL: rule of law from Kaufmann et al. (2006), RQ: regulatory quality from Kaufmann et al. (2006).

Column (2) presents two stage least squares regression results to check the robustness of results in column (1). The instruments used in two stage least square regression are religion index, logarithm of the number of publicly listed companies in the host country, logarithm of the gross domestic product of host country, rule of law index and ethno linguistic fractionalisation index. Size variable (SZ) is negative and insignificant, implying that investors do not follow the traditional financial theory as expected and their investment decisions may be driven by other factors. This result is in accordance with Amadi (2004). Internet variable (INT) is negative and significant at one per cent level, with the coefficient increasing from −0.15 (column (1)) to −0.60 (column (2)). The result supports the view that internet provides a vast amount of financial information to investors and, accordingly, investors are better able to invest in foreign equities. This result is in accordance with Amadi (2004). Trade variable (TRAD) is negative and significant at five per cent, with the coefficient increasing from −0.01 (column (1)) to −0.03 (column (2)), supporting the view that Australian investors may be willing to invest in countries with which Australia has trade relations. Covariance variable is both economically and statistically insignificant, suggesting low diversification motives of investors. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

68


MARCH

In column (3), size variable (SZ) enters significantly at the one per cent level with a negative sign, which is contrary to theoretical consideration that as investors’ local market’s share of world market capitalisation increases, an investor would decrease their foreign investment. Foreign listing variable (FL) is negative and significant at ten per cent level, implying that the greater the share of foreign firms listed in the domestic market, the greater the visibility of foreign firms by local investors and the greater the access of foreign equities by local investors. Internet variable (INT) is negative and significant at ten per cent, implying that the internet enables investors to gather financial information on equity investment and investors can utilise this information to invest in foreign equities. Trade variable (TRAD) is negative and significant at ten per cent, implying that Australian investors prefer investing in countries with which Australia has trade relationships. Countries that trade with each other tend to learn more about each other’s culture, legal and financial environment, and current accounting practices, etc. This facilitates portfolio equity investment among the countries and consequently reduces equity home bias. The diversification variable, i.e. reward to risk ratio (RWRSK), is positive and insignificant, implying low diversification motives of Australian investors. The overall adjusted R2 increases from 0.14 (column (1)) to 0.24 (column (3)). Column (4) presents two stage least squares regression results to check the robustness of results of column (3). The instruments used in two stage least squares regression are religion index, logarithm of the number of publicly – listed companies in the host country, logarithm of the gross domestic product of host country, rule of law index and ethno linguistic fractionalisation index. Size variable (SZ) is negative and insignificant, implying that investors do not follow the traditional financial theory as expected and their investment decisions may be driven by other factors. This result is in accordance with Amadi (2004) and contrary to Ahearne et al. (2004). Foreign listing variable (FL) is negative and significant at ten per cent level, with the coefficient increasing from −0.45 (column (3)) to −0.91 (column (4)). The result supports the view that local investors weight those foreign firms in their portfolios which are listed in their domestic markets. This result is in accordance with Ahearne et al. (2004). Internet variable (INT) is negative and significant at five per cent level, with the coefficient increasing from −0.11 (column (3)) to −0.61 (column (4)). The result supports the view that investors are better able to gather information on equity investment via the internet. Based on the information obtained, investors are able to invest in foreign equities. The result is in accordance with Amadi (2004). Trade variable (TRAD) is negative and insignificant. This is contrary to Lane and Milesi-Ferreti’s (2004) result that bilateral equity investment is strongly correlated with the underlying patterns of trade in goods. Mishra and Daly (2006) state that Australia’s bias towards investing in three of the world’s developed capital markets, namely the US, UK and Japan, with some deviations from that baseline due to Australian trading patterns, may be interpreted as an extension of the home bias puzzle. Reward to risk ratio variable (RWRSK) remains positive and becomes significant as compared to column (3). In column (5), size variable (SZ) enters significantly at the five per cent level with a negative sign, which is contrary to theoretical view that home bias should increase as investors’ local market’s share of world market capitalisation increases. Foreign listing variable (FL) is negative and significant at ten per cent level, implying that local investors prefer holding of equities in foreign firms which are listed in their domestic market. (TRAD) is negative and significant at five per cent, implying that Australian investors prefer investing in countries with which Australia has trade relationships because investors are better able to learn more about each other’s culture, legal and financial environment, and current accounting practices, etc. This facilitates portfolio equity investment among the countries and consequently reduces equity home bias. Transaction cost variable (TRAN) is both statistically and economically insignificant, implying that transaction cost does not impact Australia’s equity home bias. The overall adjusted R2 is 0.25. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008


69

Column (6) presents two stage least squares regression results to check the robustness of results of column (5). The instruments used in two stage least squares regression are religion index, logarithm of the number of publicly listed companies in the host country, logarithm of the gross domestic product of host country, rule of law index and regulatory quality index. Size variable (SZ) is negative and insignificant, implying that investors do not follow the traditional financial theory as expected and their investment decisions may be driven by other factors. Foreign listing variable (FL) is negative and significant at ten per cent level, with the coefficient increasing from −0.43 (column (5)) to −0.67 (column (6)). The result supports the view that foreign firms that list in local markets become more visible to local investors and, accordingly, local investors weight those firms in their portfolios. The result is in accordance with Ahearne et al. (2004). Trade variable (TRAD) is negative and significant. This is in accordance with Mishra’s (2007) result that bilateral equity investment is strongly correlated with the underlying patterns of trade in goods and services. Transaction cost variable (TRAN) is both statistically and economically insignificant. This suggests that transaction costs do not impact Australian investor’s equity investment decision. This paper employs panel regression techniques to investigate Australia’s equity home bias. The paper also employs two stage least squares panel regression techniques to check the robustness of results. Overall the results indicate that the share of the number of firms listed in domestic market and share of internet users in the total population of host country has significant impact on equity home bias. Trade linkages are found to have mixed impact on equity home bias. Results also state that country’s market share of the world market capitalisation and transaction costs do not impact Australia’s equity home bias. Investors are found to exhibit low diversification motives.

VII.

Conclusion

This paper employs IMF’s high quality CPIS dataset on cross border equity investment to investigate the determinants of equity home bias in the Australian context. On the empirical front, the paper conducts robustness tests by employing instrumental variables that are standard in the financial economics literature. The traditional studies on home bias assume that investors can hold world market portfolios. However, in a world with controlling shareholders; portfolio investors can only hold the world portfolio of shares that are not available to controlling shareholders (world float portfolio). The paper constructs float measure of home bias for the years 2001 to 2005 and explores the determinants of Australia’s equity home bias. The paper finds that the internet has significant impact on Australia’s equity home bias. The internet plays a major role in supplying financial information to investors through its financial web sites. Investors may utilise the financial information to invest abroad. The paper finds that the share of foreign firms listed in the domestic market has a significant impact on home bias. Foreign firms listing their shares in the domestic market makes foreign equity more accessible to local investors. Foreign firms listing in the domestic market alleviates a significant information cost to local investors and investors less severely underweight these firms in their portfolio. Trade links are found to have a mixed impact on home bias based on instrumental variable robustness tests. Trade alleviates certain information asymmetries in terms of familiarity with the financial and legal environment of the countries; cultural barriers, etc. Information flows © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

70


MARCH

positively affect both cross-border finance and trade. Trade in goods and trade in assets become complementary: firm managers learn about each other by trading goods and/or securities. Trading in goods market reduces informational asymmetries in the financial markets (and vice versa). The paper finds that country’s market share of the world market capitalisation and transaction costs do not impact Australia’s equity home bias. The paper also finds low diversification motives of Australian investors. The main purpose of this paper is to construct float adjusted measure of home bias, to analyse causes for the home bias in the Australian context and to derive implications from these findings for economic policy. Overall, the results indicate that information costs impact on Australia’s cross border equity holdings. This paper finds that even if policy induced barriers to equity flows have been lifted, there remains substantial economic or market inherent barriers. These barriers tend to remain relevant and to affect the way in which financial systems operate and integrate, even if economic policy has reduced regulatory barriers to entry. The market – inherent barriers due to fixed costs of market entry, including transaction costs, do not have a major impact in the Australian context.

Appendix A Countries 2001

2002

2003

2004

2005

Hong Kong Denmark Finland France Germany Greece Hungary Ireland Italy Japan South Korea Malaysia Netherlands New Zealand Norway Singapore South Africa Spain Sweden Switzerland UK US Canada Belgium Luxembourg

Austria Denmark Finland France Germany Greece Ireland Italy Japan Malaysia South Africa Spain Sweden Switzerland UK US Canada Belgium Luxembourg

Austria Brazil Hong Kong Denmark Finland France Germany Greece Hungary Ireland Italy Japan Malaysia Netherlands New Zealand Norway Singapore South Africa Spain Sweden Switzerland Thailand UK US Canada Belgium Luxembourg

Austria Hong Kong Czech Republic Denmark Finland Germany Japan South Korea Malaysia Netherlands New Zealand Norway Singapore South Africa Spain Sweden Thailand UK US Canada Luxembourg

Austria Hong Kong Czech Republic Finland Germany Japan South Korea Malaysia Netherlands Norway Singapore Thailand Turkey UK US Luxembourg

Note: The table includes only those countries for which CPIS data on countries equity investment in Australia and Australia’s equity investment in these countries is available.


2008


71

Appendix B Number of Firms in Respective Countries for Calculation of Free Float Market Capitalisation Countries Australia Austria Brazil France Canada Czech Republic Denmark Finland Chile Israel Thailand Singapore Philippines Malaysia South Korea Japan Indonesia India Hong Kong China Bangladesh Turkey Switzerland Sweden Spain Russia Romania Portugal Poland Norway Netherlands Luxembourg Lithuania Italy Ireland

Number of firms 1483 94 330 874 1441 25 172 146 198 549 353 667 198 596 1001 3830 228 706 1057 1152 5 162 378 330 158 148 51 72 189 177 146 36 41 289 56

Countries Hungary Greece Zimbabwe South Africa Morocco Mauritius Kenya UK US New Zealand Belgium Bulgaria Croatia Cyprus Germany Pakistan Sri Lanka Taiwan Egypt Argentina Colombia Ecuador Mexico Venezuela Peru Estonia Slovak Republic Malta Slovenia Iceland

Number of firms 33 303 21 363 31 7 11 1840 1130 132 208 4 4 25 451 32 50 1175 46 49 14 2 122 10 82 15 5 14 7 7

Note: Data on the number of firms is per year. Data on number of firms for Czech Republic, Bangladesh, Mauritius and Iceland correspond to market capitalisation as float adjusted market capitalisation is not available for these countries.

References Alesina, A., Devleeschauwer, A., Easterly, W., Kurlat, S. and Wacziarg, R. 2003, ‘Fractionalization’, Journal of Economic Growth, vol. 8, pp. 155–194. Amadi, A. 2004, ‘Equity Home Bias: A Disappearing Phenomenon?’, Job Market Paper, University of California, Davis USA. Ahearne, A.G., Griever, W.L. and Warnock, F.E. 2004, ‘Information Costs and Home Bias: An Analysis of US Holdings of Foreign Equities’, Journal of International Economics, vol. 62, pp. 313–336. Aviat, A. and Coeurdacier, N. 2007, ‘The Geography of Trade in Goods and Asset Holdings’, Journal of International Economics, vol. 71, pp. 22–51. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

72


MARCH

Baker, H.K., Nofsinger, J.R. and Weaver, D.G. 2002, ‘International Cross-Listing and Visibility’, Journal of Financial and Quantitative Analysis, vol. 37, no. 3, pp. 495–521. Black, F. 1974, ‘International Capital Market Equilibrium with Investment Barriers’, Journal of Financial Economics, vol. 1. Bohn, H. and Tesar, L. 1996, ‘US Equity Investment in Foreign Markets: Portfolio Rebalancing or Return Chasing’, American Economic Review, vol. 86, no. 2, pp. 77–81. Brennan, M. and Cao, H. 1997, ‘International Portfolio Investment Flows’, Journal of Finance, vol. 52, no. 2, pp. 1851–80. Chan, K., Covrig, V. and Ng, L. 2005, ‘What Determines the Domestic Bias and Foreign Bias? Evidence from Mutual Fund Equity Allocations Worldwide’, Journal of Finance, vol. LX 3, June, pp. 1495 –1534. Choe, H., Kho, B. and Stulz, R.M. 2001, ‘Do Domestic Investors Have More Valuable Information about Individual Stocks than Foreign Investors?’, NBER Working Paper, no. 8073. Cambridge, MA. Cooper, I.A. and Lessard, D.R. 1981, ‘International Capital Market Equilibrium with Deadweight Costs to Foreign Investment’, unpublished paper. Cooper, I. and Kaplanis, E. 1986, ‘Costs to Crossborder Investment and International Equity Market Equilibrium’, in J. Edwards, J. Franks, C. Mayer and S. Schaefer (eds), Recent Developments in Corporate Finance, Cambridge, Cambridge University Press. —— 1994, ‘What Explains the Home Bias in Portfolio Investment?’, Review of Financial Studies, vol. 7, pp. 45 – 60. Coval, J. and Moskowitz, T. 1999, ‘Home Bias at Home: Local Equity Preference in Domestic Portfolios’, Journal of Finance, vol. 54, pp. 2045 – 2073. —— 2001, ‘The Geography of Investment: Informed Trading and Asset Prices’, Journal of Political Economy, vol. 109, pp. 811– 841. Dahlquist, M. and Robertsson, G. 2001, ‘Direct Foreign Ownership, Institutional Investors, and Firm Characteristics’, Journal of Financial Economics, vol. 59, pp. 413 – 440. Dahlquist, M., Pinkowitz, L., Stulz, R. and Williamson, R. 2003, ‘Corporate Governance and Home Bias’, Journal of Financial and Quantitative Analysis, vol. 38, no. 1, pp. 135 –157. Domowitz, I., Glen, J. and Madhaven, A. 2001, ‘Liquidity, Volatility, and Equity Trading Costs across Countries and Over Time’, International Finance, vol. 4, no. 2, pp. 221–255. French, K. and Poterba, J. 1991, ‘Investor Diversification and International Equity Markets’, American Economic Review, Papers and Proceedings, pp. 222–226. Gehrig, T. 1993, ‘An Information Based Explanation of the Domestic Bias in International Equity Investment’, Scandinavian Journal of Economics, vol. 95, no. 1, pp. 97–109. Grinblatt, M. and Keloharju, M. 2001, ‘How Distance, Language and Culture Influence Stockholdings and Trades’, Journal of Finance, vol. 56, pp. 1053–1073. Hasan, I. and Simaan, Y. 2000, ‘A Rational Explanation for Home Country Bias’, Journal of International Money and Finance, vol. 19, pp. 331–361. Hau, H. 2001, ‘Location Matters’, Journal of Finance, vol. 56, pp. 1959–1983. Huberman, G. 2001, ‘Familiarity Breeds Investment’, Review of Financial Studies, vol. 14, no. 3, Fall, pp. 659–80. International Monetary Fund 2006, Direction of Trade Statistics, Washington DC. —— 2006, Portfolio Investment: Coordinated Portfolio Investment Survey (CPIS): Metadata, Washington DC. Kang, J.K. and Stulz, R. 1997, ‘Why is There Home Bias? An Analysis of Foreign Equity Ownership in Japan’, Journal of Financial Economics, vol. 46, no. 1, pp. 3–28. Karolyi, G.A. and Stulz, R.M. 2003, ‘Are Financial Assets Priced Locally or Globally?’, in G.M. Constantinidies, M. Harris and R. Stulz (eds), The Handbook of the Economic Finance, New York NY, North Holland, pp. 975 –1020. Kaufmann, D., Kraay, A. and Zoido-Lobaton, P. 2006, ‘Governance Matters V: Governance Indicators for 1996 –2005’, World Bank Policy Research Working Paper Series, no. 4012. Kho, B.C., Stulz, R.M. and Warnock, F.E. 2006, ‘Financial Globalization, Governance and the Evolution of the Home Bias’, NBER Working Paper, no. 12389. Kim, W. and Wei, Shang-Jin 2002, ‘Foreign Portfolio Investors Before and During a Crisis’, Journal of International Economics, vol. 56, pp. 77–96. © 2008 The Author Journal compilation © 2008 Blackwell Publishing Ltd / University of Adelaide and Flinders University

2008


73

Lewis, K. 1999, ‘Trying to Explain Home Bias in Equities and Consumption’, Journal of Economics Literature, vol. 37, no. 2, pp. 571–608. La Porta, R., Lopez-de-Silanes, F. and Sheifer, A. 1999, ‘Corporate Ownership Around the World’, Journal of Finance, vol. 54, pp. 471–517. Lane, P.R. and Milesi-Ferreti, G.M. 2004, ‘International Investment Patterns’, IISC Discussion Paper, no. 24. Lintner, J. 1965, ‘The Valuation of Risky Assets and the Selection of Risky Investment in Stock Portfolio and Capital Budgets’, Review of Economics and Statistics, vol. 47, pp. 103–124. Li, S., Hamid, F. and Isabel, K.Y. 2004, ‘The Determinants of International Portfolio Holdings and Home Bias’, IMF Working Paper, no. WP/04/34. Mauro, P. 1995, ‘Corruption and Growth’, Quarterly Journal of Economics, vol. 110, no. 3, August, pp. 681–712. Merton, R. 1987, ‘A Simple Model of Capital Market Equilibrium with Incomplete Information’, Journal of Finance, vol. 42, pp. 483–510. Mishra, A.V. 2007, ‘International Investment Patterns: Evidence Using a New Dataset’, Research in International Business and Finance, vol. 21, no. 2, pp. 342–360. Mishra, A. and Daly, K. 2007, ‘Effect of Quality of Institutions on Outward Foreign Direct Investment’, Journal of Trade and Economic Development, vol. 16, no. 2, June, pp. 231–244. —— 2006, ‘Where Do Australians Invest?’, Australian Economic Review, vol. 39, no. 1, pp. 47–59. Portes, R. and Rey, H. 2005, ‘The Determinants of Cross-border Equity Flows’, Journal of International Economics, vol. 65, no. 2, March, 269 –296. Portes, R., Rey, H. and Oh., Y. 2001, ‘Information and Capital Flows: The Determinants of Transactions in Financial Assets’, European Economic Review, vol. 45, May, pp. 783–96. Sarkissian, S. and Schill, M.J. 2004, ‘The Overseas Listing Decision: New Evidence of Proximity Preference’, Review of Financial Studies, vol. 17, no. 3, pp. 769–809. Sharpe, W. 1964, ‘Capital Asset Prices: A Theory of Market Equilibrium under the Condition of Risk’, Journal of Finance, vol. 19, pp. 425 – 442. Stulz, R. 1981, ‘On the Effects of Barriers to International Investment’, Journal of Finance, vol. 36, pp. 923–934. Tesar, L. and Werner, I.M. 1995, ‘Home Bias and High Turnover’, Journal of International Money and Finance, vol. 14, pp. 467– 493. Warnock, F. 2002, ‘Home Bias and High Turnover Reconsidered’, Journal of International Money and Finance, vol. 21, pp. 795–805. World Bank 2005, World Development Indicators, Washington DC, World Bank.
