An Investigation into the Benefits of Investing in Chinese Multinational Companies

Abstract

We provide a unique longitudinal analysis into firm-level multinationality of Chinese firms between 2002 and 2012 using two measures of multinationality—the Aggarwal, Berrill, Hutson and Kearney (2011, International Business Review, 20, 557–577) system and foreign sales as a percentage of total sales. We find that Chinese firms have low levels of multinationality with most foreign sales in the triad regions of Asia, Europe and North America. We use mean variance spanning tests to investigate the benefits from investing in Chinese multinational corporations (MNCs). We find that Chinese MNCs offer little diversification benefits to domestic investors and firms with greater multinationality levels do not result in greater benefits.

Keywords

MNCs international portfolio diversification mean variance spanning

1. Introduction

The literature defines internationalisation as a process through which a firm increases its level of involvement in foreign markets (Welch & Luostarinen, 1988), and traditionally considers it as a series of events that take place over time (Johanson & Vahlne, 1990; Leonidou & Katsikeas, 1996). The existing theories of firm-level internationalisation, together with the many patterns of internationalisation that are traced by firms of different age, size, industry and home country, suggest that there is no standard or optimal path of international expansion (Buckley & Chapman, 1997; Fillis, 2001).Whether firms internationalise slowly accor-ding to one or more of the stages theories of internationalisation, or rapidly according to International New Venture Theory, the common dimension is time. It seems surprising, therefore, that there have been so few longitudinal studies of firm-level internationalisation. Contractor (2007) and Hennart (2007) suggest that international business researchers should diversify away from an almost exclusive reliance on cross-sectional data and embrace longitudinal studies as the best approach to achieve a more complete understanding of the evolution of the multinational corporation (MNC). Glaum and Oesterle (2007) advocate the benefits of a longitudinal approach on the relation between firm-level multinationality and performance, and Brouthers and Hennart (2007) and Canabal and White (2008) suggest that longitudinal data should be used in studies of entry mode. More recently, Casillas and Acedo (2013) call for the introduction of time into studies of the internationalisation process.

We conduct a unique longitudinal analysis on the patterns of internationalisation for 300 Chinese firms over an 11-year period from 2002 (when China joined the World Trade Organisation [WTO]) to 2012. We use two measures of internationalisation—the classification scheme for firm-level multinationality introduced by Aggarwal et al. (2011) (hereafter ABHK) and foreign sales as a percentage of total sales. In ABHK’s model, firms are classified on the basis of the breadth of international engagement across six regions that encompass all countries of the world: Africa, Asia, Europe, North America, Oceania and South America. We classify firms into seven categories ranging from domestic to global, according to the location of their sales. We test the robustness of this technique using foreign sales as a percentage of total sales. We find that the two measures of internationalisation are not always consistent with each other, and capture different aspects of the degree of internationalisation of a firm. A high level of foreign sales, for example, does not necessarily mean that firms are operating across a wide geographic scope. We categorise firms using each measure in each year and examine how they change over the 11-year sample period. This gives us a unique perspective on changing patterns of internationalisation over time.

The benefits of international portfolio diversification have long been highlighted throughout the literature (Driessen & Laeven, 2007; Levy & Sarnat, 1970; Solnik, 1974). Despite these benefits, a home-biased attitude to international portfolio investment continues to exist (Ahearne, Griever, & Warnock, 2004; Suh, 2005). Possible explanations include transaction costs, taxes, information asymmetries, currency risk, legal restrictions, political risk and other controls. One explanation is that investors may be able to indirectly gain foreign exposure by investing in domestically traded products (Antoniou, Louse, & Paudyal, 2010; Errunza, Hogan, & Hung, 1999). A subset of this literature investigates the role that MNCs have to play in this regard. We investigate the international portfolio diversification benefits from investing in various categories of Chinese firms, using weekly data and mean variance spanning (MVS) tests. We also break our sample into three sub-periods—pre-crisis period, crisis period and post-crisis period.

We make several contributions to the literature. First, we provide a unique longitudinal analysis into firm-level multinationality in China. This is of particular relevance as most studies on firm-level multinationality focus on developed markets, mainly the USA, at one point in time. To date, comparatively less work has been done for emerging markets such as China. China represents an interesting market for analysis given its transfer from an autarkic economy to an advocate of globalisation and its well-documented economic growth. Second, we analyse the home-based diversification effect of MNCs from an emerging market perspective and test to what extent Chinese investors can reap the benefits of internationalisation by investing in home-based MNCs in comparison with direct international portfolio diversification. Institutional investors in countries such as China are often restricted by their governments to invest only in domestic markets (Driessen & Laeven, 2007) and can only access limited benefits by direct investment in foreign stock markets. China’s regulations allow only Qualified Domestic Institutional Investors (QDII) to invest directly in foreign markets. Therefore, this issue is of particular importance to China.

Our main findings are as follows. We find that Chinese firms tend to have low levels of multinationality using both of our measures and we find little evidence of increasing levels of multinationality over time. We find that the USA and EU are the largest exporting targets of Chinese firms, but sales to these two regions declined in recent years. In contrast, domestic sales and sales to developing regions such as Asia and Africa increased. Our MVS tests show that there are benefits to international portfolio diversification for Chinese investors, but Chinese MNCs do not yield significant benefits. Firms with lower levels of multinationality provide greater diversification benefits, which seems counter-intuitive. The best performing groups of firms are those with sales in two geographical regions (classified as T2) and foreign sales less than 5 per cent. These small benefits are reduced during the crisis period. Other categories of firms yield little diversification benefits.

The remainder of this article is structured as follows. The second section provides a review of the literature. The third section presents the data and methodology. The fourth section reports the empirical results. Finally, the fifth section concludes.

2. Literature Review

Our analysis contributes to two literatures—the regionalisation versus globalisation debate on firm-level activities and the topic on whether MNCs provide international portfolio diversification benefits. The extent to which firms are regional or global in their activities is a source of much debate within the literature. Although many international business scholars such as Yip (2002) and Govindarajan and Gupta (2008) argue that global business strategy is paramount, others such as Ghemawat (2001, 2003) argue the case for semi-global strategy, pointing to escalating costs of internationalising over greater geographical and cultural distances. Doremus, Keller, Pauly and Reich (1998) argue that we have not achieved anything remotely close to full globalisation, that state sovereignty remains strong and that the world’s largest MNCs retain a national and regional focus.

Rugman (2000, 2003, 2005), Rugman and Brain (2003), Rugman and Girod (2003), Rugman and Hodgetts (2001), Rugman and Verbeke (2003, 2004, 2007, 2008), Collinson and Rugman (2008) and Rugman and Oh (2010) argue the case for the regional dimension in international business and strategy. Rugman and his co-authors base their analysis on the triad regions of North America, Europe and Asia-Pacific (Ohmae, 1985; Rugman, 2003). They conclude that most of the world’s largest MNCs are regional rather than global, that globalisation is a myth and that regional rather than global strategy is paramount in international business. The evidence assembled by Rugman and his co-authors in favour of regionalisation rather than globalisation of the world’s largest firms has been scrutinised by Aharoni (2006), Osegowitsch and Sammartino (2007, 2008), Dunning, Fujita and Yakova (2007), Asmussen (2008) and Berrill (2015). These researchers have introduced refinements to the data analysis to show that the evidence in favour of regionalisation is not overwhelming, and they have questioned the extent to which it implies that global strategy is a myth.

Furthermore, the issue on whether MNCs indirectly provide the benefits from portfolio diversification remains unresolved in the literature. Early research by Hughes, Logue and Sweeney (1975), Agmon and Lessard (1977), Mikhail and Shawky (1979) and Logue (1982) concluded that investing in MNCs does indeed yield international diversification benefits, and more recent work by Errunza et al. (1999), Cai and Warnock (2004), Antoniou et al. (2010) and Berrill and Kearney (2010) supports these earlier findings. Errunza et al. (1999), for example, find that MNCs can be a part of domestic portfolio that is able to mimic foreign index returns for US investors. Berrill and Kearney (2010) conclude that MNCs can provide diversification benefits for all G7 countries. The issue is far from settled, however, because work by Jacquillat and Solnik (1978), Senchack and Beedles (1980), Brewer (1981), Fatemi (1984), Michel and Shaked (1986), Kim and Lyn (1990), Mathur, Singh and Gleason (2001), Wright and McCarthy (2002), Salehizadeh (2003) and Rowland and Tesar (2004) has all found that investing in MNCs does not yield significant international diversification benefits. Wright and McCarthy (2002) argue no benefits are available for Australian investors. Rowland and Tesar (2004) show the result differs in G7 countries and only Germany and the USA see substantial diversification benefits provided by MNCs. Thus, the issue is far from being settled.

This lack of agreement may be due to the different methods used to create MNC samples in the literature. For example, Michel and Shaked (1986) categorise MNCs as foreign sales above 20 per cent of revenue and direct capital investment at least six countries outside the USA. Mathur et al. (2001) define MNCs as companies that have foreign assets and sales. Errunza et al. (1999) use the 30 largest US companies in the Fortune 100 list, making the implicit assumption that large firms must be multinational. Aggarwal et al. (2011) highlight this issue and propose a classification system for firms which we apply in our analysis.

3. Data and Methodology

Data

Our sample data contain the 300 constituent firms from the China Shanghai Shenzhen 300 Index (Hushen 300 Index). This index includes stocks listed in China’s two main exchanges, Shanghai and Shenzhen. We gather data on the geographical breakdown of each constituent firm’s sales in this index for each year between 2002 and 2012 (11 years) from Datastream. We also obtain the percentage of foreign sales of each firm in each year and weekly closing share price data from 4 January 2002 to 28 December 2012. We use the 3-month deposit rate as the risk-free rate of return¹. Weekly data on the MSCI World Index, the S&P500 Index and the Chinese Hushen 300 Index are also used.

We first categorise each firm in terms of its internationalisation using the Aggarwal et al. (2011) (ABHK) classification system. Following ABHK, we divide the world into six regions: Africa, Asia, Europe, North America, Oceania and South America. We give each firm a score in each year based on the location of its sales. If a firm has no foreign sales, it is given a score of 0 (domestic); if it has sales outside China but only in Asia, it is given a score of 1 which we classify as being regional; if it has sales in Asia plus one other of our six regions, it receives a score of 2 (T2, trans-regional 2); if it has sales in Asia plus two other continents, it receives a score of 3 (T3, trans-regional 3) and so on for T4 and T5 to a maximum score of 6 (global), which indicates that a firm has sales in all six regions. A more detailed description of this classification system is contained in Aggarwal et al. (2011). In addition to the ABHK model, we use foreign sales as a percentage of total sales as an alternative measure of internationalisation.

We believe that the use of these two measures of internationalisation provides a robust approach to capturing the various divergent aspects of firm-level internationalisation. Our first measure, the ABHK model, captures the breadth of internationalisation but takes no account of the level of internationalisation. A firm is classified as T5 if it has sales in five regions of the world, but the percentage of foreign sales may differ significantly between firms. For example, in 2011, CSG Holding is classified as T5 using the ABHK model but has only 23 per cent foreign sales. China CSSC Holding is also classified as T5 and has 80 per cent foreign sales. The use of an additional measure will help to distinguish between these two firms. The percentage of foreign sales gives an accurate measure of the total level of multinationality but does not give any information about where those sales occur. A Chinese firm, for example, may have a high percentage of foreign sales but those sales may be spread across many countries or all occur in a neighbouring country. For example, in 2011, Yanzhou Coal Mining has 19 per cent foreign sales spread across Australia, Canada and ‘rest of the world’. Huaneng Power has 16 per cent foreign sales but they all occur in Singapore. The use of two measures of internationalisation helps distinguish between these two firms and provide measures of both the breadth and total level of multinationality.

3.1. Mean Variance Spanning Tests

Several methods are available to measure the extent to which MNCs provide international portfolio diversification benefits. These include using the international market model to investigate the influence of domestic and foreign market indexes on individual shares (Agmon & Lessard, 1977; Brewer, 1981; Hughes et al., 1975); comparing the risk-adjusted performance of MNCs and domestic firms; comparing firms on the basis of returns, standard deviations, betas, coefficient of variation and performance measures, such as the Sharpe, Treynor and Jensen measures (Fatemi, 1984; Jacquillat & Solnik, 1978; Michel & Shaked, 1986; Mikhail & Shawky, 1979; Senchack & Beedles, 1980); and more recently, MVS tests (Berrill & Kearney, 2010; Errunza et al., 1999; Rowland & Tesar, 2004).

Following the MVS methodology of Huberman and Kandel (1987), De Roon and Nijman (2001) and Kan and Zhou (2012), we consider a set of K benchmark assets and N test assets, and we investigate whether, conditional on the K benchmark assets, the addition of the N test assets can shift the mean variance efficient frontier. Alternatively, conditional on the K+N benchmark and test assets, can the subset of K benchmark assets yield the same diversification benefits? In other words, we are interested in whether the K benchmark assets ‘span’ the extended set of K+N assets. To do this, we define R_1,t as the K×1 vector of returns on the K benchmark assets and R_2,t as the N×1 returns on the N test assets at time t, and we combine R_1,t and R_2,t in the K+N vector $R_{t} = [R_{1, t}^{'}, R_{2, t}^{'}]^{'} .$ The MVS test proceeds by regressing the N test asset returns on the K benchmark returns as follows:

R_{2, t} = α + β R_{1, t} + ε_{t}

\begin{matrix} w i t h ε_{t} ~ N (0, Σ), α = E [R_{2, t}] - β E [R_{1, t}] = μ_{2} - β μ_{1}, a n d \\ β = V_{21} V_{11}^{- 1} . \end{matrix}

(1)

We estimate Equation (1) using ordinary least squares. The null hypothesis is that the benchmark portfolio spans the extended portfolio comprising the benchmark assets and the test assets, implying that the mean variance frontiers coincide at all points. Testing whether the K benchmark assets span the broader set of K+N assets amounts to testing the joint hypothesis that α = 0 and β = 1. If this hypothesis is upheld, it implies that for every test asset, we can obtain a portfolio of the K benchmark assets that has the same expected return (because α = 0 and β = 1) and a lower variance (because R_1,t and ε_t are uncorrelated while Var(ε_t) is positive definite).

We also perform a two-step Wald test (Kan & Zhou, 2012), whereby we first test whether α = 0, and we then test whether β = 1 conditional on α = 0. If we reject the null hypothesis of spanning due to the first test, the tangency portfolios are different, and if we reject due to the second test, the global minimum variance portfolios are different. If the null hypothesis of spanning is rejected, we measure the economic significance of the diversification benefits by changes in the Sharpe ratios between the K benchmark assets and the K + N assets. Different Sharpe ratios indicate that investors can improve their risk-return trade-offs by investing in the additional assets.

In order to test the robustness of our results to the financial crises, we divide our sample period into three sub-periods.² We use two methods to establish sub-periods for our analysis—the Chow (1960) test and the Bai–Perron (1998) test. Both tests confirm a break point in the data on 19 October 2007. The Chow test further indicates a break point on 7 August 2009. Therefore, we divide our sample period into the following three subsamples—pre-crisis (4 January 2002 to 19 October 2007), crisis (19 October 2007 to 7 August 2009) and post-crisis (7 August 2009 to 31 December 2012) and repeat our analysis for each sub-period.

4. Results

4.1. Longitudinal Analysis of Firm-level Multinationality

We begin by analysing changing patterns of multinationality for our sample firms between 2002 and 2012. We use two measures of multinationality—the multinational classification system proposed by Aggarwal et al. (2011) and foreign sales as a per cent of total sales. Table 1 shows that the average foreign sales for the 11-year sample period were 11.98 per cent. It is more than quadrupled from 4.67 per cent in 2002 to 18.32 per cent in 2012, with the largest increases in 2003 (from 4.67% to 12.83%) and 2012 (from 11.40% to 18.32%). These increases are likely the result of China’s decision to join the WTO in 2002 and measures introduced to stimulate foreign trade in 2012.³ The percentage of foreign sales fell in 2007 and 2008, likely as a result of the financial crisis. However, despite the overall increase in average foreign sales percentages, the majority of sales continue to lie in the domestic market, with domestic sales percentages ranging from 96.37 per cent in 2002 to 83.14 per cent in 2012.

Table 1.
Foreign Sales Analysis

Year Sample Size Domestic
Asia
Other Foreign

China Internal
Adjust HK
Macao
Taiwan Other Asia North America Europe Australia Africa South America Other % FS

Panel A: Full Sample

2002 49 96.37% −1.03% 0.43% 0.52% 0.61% 1.02% 0.01% 0.00% 0.00% 2.08% 4.67%

2003 59 87.46% −0.29% 1.17% 1.08% 1.49% 2.31% 0.01% 0.00% 0.00% 6.77% 12.83%

2004 68 88.05% −0.42% 0.63% 1.64% 2.23% 3.20% 0.01% 0.34% 0.00% 4.32% 12.37%

2005 99 86.14% −0.39% 0.51% 2.53% 1.20% 1.46% 0.01% 0.33% 0.00% 8.22% 14.24%

2006 109 84.50% −0.53% 4.53% 2.09% 0.76% 0.75% 0.00% 0.15% 0.00% 7.74% 16.03%

2007 114 89.72% −0.33% 2.42% 1.05% 0.46% 1.23% 0.00% 0.15% 0.00% 5.31% 10.61%

2008 118 90.66% −0.30% 1.35% 1.44% 0.18% 0.69% 0.00% 0.19% 0.00% 5.79% 9.64%

2009 125 89.38% −0.02% 1.03% 0.94% 0.14% 0.41% 0.00% 0.12% 0.00% 7.99% 10.64%

2010 126 88.98% −0.02% 0.86% 0.77% 0.16% 0.48% 0.00% 0.14% 0.00% 8.63% 11.04%

2011 125 88.83% −0.23% 0.73% 0.78% 0.15% 0.30% 0.00% 0.18% 0.00% 9.25% 11.40%

2012 158 83.14% −1.47% 0.45% 0.51% 0.10% 0.21% 0.02% 0.06% 0.00% 16.97% 18.32%

AVG 88.48% −0.46% 1.28% 1.21% 0.68% 1.10% 0.01% 0.15% 0.00% 7.55% 11.98%

Panel B: 33 Firm Sample

2002 33 88.85% −3.82% 1.55% 1.40% 1.77% 2.89% 0.00% 0.00% 0.00% 1.40% 14.97%

2003 33 86.94% −0.51% 1.11% 0.98% 2.16% 3.46% 0.00% 0.00% 0.00% 0.98% 13.57%

2004 33 83.15% 0.00% 1.10% 1.90% 2.75% 5.39% 0.01% 0.59% 0.00% 1.90% 16.85%

2005 33 79.94% −0.01% 1.06% 5.69% 3.06% 4.13% 0.00% 0.95% 0.00% 5.69% 20.07%

2006 33 77.17% 0.00% 0.84% 5.67% 2.24% 3.08% 0.00% 0.66% 0.00% 5.67% 22.83%

2007 33 79.58% −0.88% 0.52% 3.75% 2.34% 4.69% 0.00% 1.05% 0.00% 3.75% 21.30%

2008 33 81.65% −0.18% 0.38% 5.39% 0.48% 2.13% 0.00% 1.44% 0.00% 5.39% 18.53%

2009 33 84.53% −0.12% 0.29% 4.70% 0.54% 1.69% 0.00% 1.01% 0.00% 4.70% 15.59%

2010 33 84.38% −0.37% 0.45% 3.94% 0.72% 1.82% 0.00% 1.22% 0.00% 3.94% 15.99%

2011 33 84.49% −0.37% 0.36% 4.74% 0.70% 0.95% 0.00% 1.02% 0.00% 4.74% 15.88%

2012 33 84.65% −0.48% 0.36% 4.49% 0.51% 0.82% 0.00% 0.77% 0.00% 4.49% 15.83%

AVG 83.21% −0.61% 0.73% 3.88% 1.57% 2.82% 0.00% 0.79% 0.00% 3.88% 17.40%

Year	Sample Size	Domestic	Asia	Other Foreign
Panel A: Full Sample
2002	49	96.37%	−1.03%	0.43%	0.52%	0.61%	1.02%	0.01%	0.00%	0.00%	2.08%	4.67%
2003	59	87.46%	−0.29%	1.17%	1.08%	1.49%	2.31%	0.01%	0.00%	0.00%	6.77%	12.83%
2004	68	88.05%	−0.42%	0.63%	1.64%	2.23%	3.20%	0.01%	0.34%	0.00%	4.32%	12.37%
2005	99	86.14%	−0.39%	0.51%	2.53%	1.20%	1.46%	0.01%	0.33%	0.00%	8.22%	14.24%
2006	109	84.50%	−0.53%	4.53%	2.09%	0.76%	0.75%	0.00%	0.15%	0.00%	7.74%	16.03%
2007	114	89.72%	−0.33%	2.42%	1.05%	0.46%	1.23%	0.00%	0.15%	0.00%	5.31%	10.61%
2008	118	90.66%	−0.30%	1.35%	1.44%	0.18%	0.69%	0.00%	0.19%	0.00%	5.79%	9.64%
2009	125	89.38%	−0.02%	1.03%	0.94%	0.14%	0.41%	0.00%	0.12%	0.00%	7.99%	10.64%
2010	126	88.98%	−0.02%	0.86%	0.77%	0.16%	0.48%	0.00%	0.14%	0.00%	8.63%	11.04%
2011	125	88.83%	−0.23%	0.73%	0.78%	0.15%	0.30%	0.00%	0.18%	0.00%	9.25%	11.40%
2012	158	83.14%	−1.47%	0.45%	0.51%	0.10%	0.21%	0.02%	0.06%	0.00%	16.97%	18.32%
AVG		88.48%	−0.46%	1.28%	1.21%	0.68%	1.10%	0.01%	0.15%	0.00%	7.55%	11.98%
Panel B: 33 Firm Sample
2002	33	88.85%	−3.82%	1.55%	1.40%	1.77%	2.89%	0.00%	0.00%	0.00%	1.40%	14.97%
2003	33	86.94%	−0.51%	1.11%	0.98%	2.16%	3.46%	0.00%	0.00%	0.00%	0.98%	13.57%
2004	33	83.15%	0.00%	1.10%	1.90%	2.75%	5.39%	0.01%	0.59%	0.00%	1.90%	16.85%
2005	33	79.94%	−0.01%	1.06%	5.69%	3.06%	4.13%	0.00%	0.95%	0.00%	5.69%	20.07%
2006	33	77.17%	0.00%	0.84%	5.67%	2.24%	3.08%	0.00%	0.66%	0.00%	5.67%	22.83%
2007	33	79.58%	−0.88%	0.52%	3.75%	2.34%	4.69%	0.00%	1.05%	0.00%	3.75%	21.30%
2008	33	81.65%	−0.18%	0.38%	5.39%	0.48%	2.13%	0.00%	1.44%	0.00%	5.39%	18.53%
2009	33	84.53%	−0.12%	0.29%	4.70%	0.54%	1.69%	0.00%	1.01%	0.00%	4.70%	15.59%
2010	33	84.38%	−0.37%	0.45%	3.94%	0.72%	1.82%	0.00%	1.22%	0.00%	3.94%	15.99%
2011	33	84.49%	−0.37%	0.36%	4.74%	0.70%	0.95%	0.00%	1.02%	0.00%	4.74%	15.88%
2012	33	84.65%	−0.48%	0.36%	4.49%	0.51%	0.82%	0.00%	0.77%	0.00%	4.49%	15.83%
AVG		83.21%	−0.61%	0.73%	3.88%	1.57%	2.82%	0.00%	0.79%	0.00%	3.88%	17.40%

Source: Authors’ own calculations.

Notes: This table shows the percentage sales to each of the six regions of the world and the total per cent foreign sales figures for our sample firms. Panel A includes all firms with at least 1 year’s foreign sales disclosure from 2002 to 2012. Panel B includes only firms that disclose foreign sales figures in each year from 2002 to 2012.

The geographical breakdown of foreign sales in Table 1 shows that Asia accounts for the largest proportion of foreign sales (2.49% on average), followed by Europe (1.10%) and North America (0.68%). Africa accounts for 0.15 per cent of foreign sales on average, Australia for just 0.01 per cent while no foreign sales are reported in South America in any year. The proportion of foreign sales categorised as ‘Other’ increased dramatically from 2.08 per cent in 2002 to 16.97 per cent in 2012⁴. Foreign sales to Europe and North America show similar trends with both increasing between 2002 and 2004, and hitting a maximum value in 2004 (3.20% for Europe and 2.23% for North America) but sales to both regions have fallen since then and in 2012 stand at 0.21 per cent for Europe and 0.10 per cent for North America.

In order to test the robustness of our results, we create a second sample of firms including only those that report a geographical sales breakdown in each year of our sample—33 firms. Results are presented in Panel B of Table 1. For these 33 firms, the average foreign sales figure is 17.40 per cent. The average percentage of domestic sales is 83.21 per cent. Both samples show that the subprime crisis had a negative effect on China’s foreign sales levels. The two sample groups show similar trends until 2012, where sample 1 shows a dramatic increase that is not matched in sample 2. This may indicate that the increase in China’s foreign sales largely results from the rising number of MNCs rather than deepening in foreign sales of existing MNCs. The results from sample 2 again show that most foreign sales lie in Asia, Europe and North America. The last decade witnessed a stable increase in sales to Africa each year. Moreover, the results from this smaller sample also show that sales to Europe are higher than those to North America in each year, and sales to both regions increased rapidly at the start of the sample period but later slowed down and fell in more recent years.

We next use the Aggarwal et al. (2011) system to categorise all firms into seven categories in Table 2, ranging from domestic to global. The number of domestic firms doubled from 30 in 2002 to 60 in 2005 but fell to 52 in 2012. The percentage of domestic firms increased in 2004 and 2005 but fell in every other year. The percentage of domestic firms almost halved over the sample period from 61.22 per cent in 2002 to 32.91 per cent in 2012. No firms were categorised as regional until 2006. In 2006, two firms were classified as regional and this figure increased to five (3.16%) in 2012. This finding is in line with previous research which finds little evidence in favour of the regionalisation of firms (refer, e.g., Rugman & Verbeke, 2004, 2008).

Table 2.

ABHK Classification of Firms

	Domestic (D)			Regional (R)			Trans-regional (T2)			Trans-regional (T3)
	No. of firms	%	AVG FS %	No. of firms	%	AVG FS %	No. of firms	%	AVG FS %	No. of firms	%	AVG FS %
2002	30	61.22%	0.00%	0	0.00%	N/A	16	32.65%	20.62%	1	2.04%	6.94%
2003	33	55.93%	0.00%	0	0.00%	N/A	23	38.98%	18.77%	0	0.00%	N/A
2004	39	57.35%	0.00%	0	0.00%	N/A	25	36.76%	17.29%	0	0.00%	N/A
2005	60	60.61%	0.00%	0	0.00%	N/A	35	35.35%	25.29%	0	0.00%	N/A
2006	63	57.80%	0.00%	2	1.83%	1.57%	40	36.70%	20.41%	1	0.92%	44.42%
2007	60	52.63%	0.00%	1	0.88%	0.19%	45	39.47%	10.88%	1	0.88%	32.23%
2008	53	44.92%	0.00%	2	1.69%	9.67%	56	47.46%	10.04%	2	1.69%	58.34%
2009	52	41.60%	0.00%	3	2.40%	10.77%	62	49.60%	11.50%	4	3.20%	38.60%
2010	52	41.27%	0.00%	4	3.17%	7.87%	62	49.21%	11.97%	4	3.17%	24.13%
2011	51	40.80%	0.00%	4	3.20%	5.84%	59	47.20%	13.22%	5	4.00%	13.61%
2012	52	32.91%	0.00%	5	3.16%	5.41%	92	58.23%	21.19%	2	1.27%	26.01%
AVG	49.55	47.39%	0.00%	1.91	1.83%	6.49%	46.82	44.78%	15.11%	1.82	1.74%	23.91%

	Trans-regional (T4)			Trans-regional (T5)			Global (G)			Total Firms	AVG Value
	No. of firms	%	AVG FS %	No .of firms	%	AVG FS %	No. of firms	%	AVG FS %	Total Firms	AVG Value
2002	2	4.08%	70.18%	0	0.00%	N/A	0	0.00%	N/A	49	0.88
2003	3	5.08%	63.82%	0	0.00%	N/A	0	0.00%	N/A	59	0.98
2004	4	5.88%	51.38%	0	0.00%	N/A	0	0.00%	N/A	68	0.97
2005	4	4.04%	67.25%	0	0.00%	N/A	0	0.00%	N/A	99	0.87
2006	3	2.75%	70.58%	0	0.00%	N/A	0	0.00%	N/A	109	0.89
2007	6	5.26%	59.94%	1	0.88%	72.15%	0	0.00%	N/A	114	1.08
2008	4	3.39%	43.72%	1	0.85%	80.17%	0	0.00%	N/A	118	1.19
2009	3	2.40%	42.37%	1	0.80%	78.76%	0	0.00%	N/A	125	1.25
2010	1	0.79%	84.65%	2	1.59%	67.85%	1	0.79%	54.18%	126	1.27
2011	2	1.60%	38.83%	3	2.40%	60.03%	1	0.80%	54.21%	125	1.33
2012	5	3.16%	37.90%	1	0.63%	9.17%	1	0.63%	53.03%	158	1.43
AVG	3.36	3.22%	52.95%	0.82	0.78%	66.13%	0.27	0.26%	53.79%	104.6	1.15

Source: Authors’ own calculations.

Notes: This table shows the number and proportion of firms in each category ranging from domestic to global using the ABHK classification system. It also shows the average foreign sales percentage for firms in each category in each year. For example, in 2012, 92 firms (58.23%) were categorised as T2. These firms had an average of 21.19 per cent foreign sales.

Most firms in our sample are classified as T2, with the number and percentage of firms in this category increasing each year. However, for all other categories (R, T3, T4, T5, G), the figures and percentages are extremely low with no obvious increase during the sample period. This suggests a pattern of internationalisation whereby Chinese firms tend to expand into one other geographical region but tend not to expand any further. It may suggest that Chinese firms are at an early stage of international expansion. The classification shows a lack of multinationality of China’s firms. According to Xing (2003), this is likely because of high state involvement in many firms and the fact that many macro-management regulations hinder the progress of globalisation. Firms face strict controls in raising funds for foreign investing, which requires layers of approval. We conclude that the increase in China’s foreign sales level is mostly ascribed to the rising number of low-level MNCs (T2) rather than increasing foreign sales for existing firms. Results from our smaller 33 firm samples support this hypothesis.

4.2. Mean Variance Spanning Tests

We use several MVS tests to investigate the extent to which Chinese investors can gain the benefits of internationalisation via investing in Chinese firms. We first test the benefits of international diversification using the Hushen 300 Index as the benchmark portfolio and the MSCI World Index as the extended set. The MVS tests confirm that benefits exist at the 5 per cent significance level, although adding the MSCI World Index into Chinese assets does not yield a higher Sharpe ratio. When the sample period is broken down into our three subsamples, results show that there are benefits to international diversification in the pre- and post-crisis period (Sharpe ratio increases of 5.38% and 143.35%, respectively), but no benefits during the crisis period.

We focus our analysis on an investigation into the extent that Chinese investors can gain international diversification benefits by investing in home-based MNCs. In our tests, Chinese domestic firms form the benchmark portfolio with various categories of multinational firms (regional, trans-regional and global) forming the extended sets. We estimate Equation (1) and test the robustness of our results using both market value and equally weighted indices. We do not allow short sales in our analysis given that China’s Law of Securities Regulation does not allow the short selling of shares⁵. Results in Panel A of Table 3 show that Chinese MNCs yield almost no additional benefits to domestic investors. For the market value weighted regressions, the null hypothesis of spanning is rejected for all time periods, indicating that portfolios of MNCs shift the efficient frontier of domestic firms. However, there is almost no improvement in the Sharpe ratios with the exception of a 0.57 per cent increase in the pre-crisis period. Using the equally weighted data, we fail to reject the hypothesis of spanning for all periods. Thus, there is no benefit to adding MNCs to a portfolio of purely domestic firms for Chinese investors. These results support the findings of Wright and McCarthy (2002), Salehizadeh (2003) and Rowland and Tesar (2004) who all find that investing in MNCs does not yield diversification benefits albeit in developed markets.

Although MNCs as a whole do not provide diversification benefits, we investigate whether this result applies to all categories of MNC. Given the dominance of firms categorised as T2 (44.78%—refer Table 2) in our sample, we further segregate this category into three levels according to their foreign sales percentages—T2 firms with foreign sales less than 5 per cent, between 5 per cent and 15 per cent, and above 15 per cent. As there are significantly less firms for T3, T4, T5 and global categories, we combine these firms to make one portfolio. Regional firms are not included as the number is extremely low. Table 4 (test 1) lists the number of firms in each category in each year. We create portfolios using a maximum of 10 firms and all firms are included for categories with less than 10 firms. This eliminates any extra diversification benefits created by larger portfolios. The results are presented in Panels B and C of Table 3.

We reject the null hypothesis of spanning in all tests, implying that diversification benefits exist in all cases (refer Table 3). For T2 firms with less than 5 per cent foreign sales, there is a slight increase in the Sharpe ratio for the entire sample period but no benefits in the sub-periods (using market capitalised weighted portfolios). Results using equally weighted portfolios show that MNCs do provide diversification benefits with an overall Sharpe ratio increase of 2.88 per cent. The remaining tests show similar results. Results for the full sample period show no change in the Sharpe ratio using market value or equally weighted data. Several sub-periods show additional benefits from investing in various categories of MNCs, but these benefits are not consistent. The T2 (<5% foreign sales) category shows relatively higher potential in providing extra benefits (0.03% for weighted average and 2.88% for equally weighted).

We test the robustness of our results by categorising firms into three categories based on their foreign sales percentages—0–5 per cent, 5–15 per cent and greater than 15 per cent. Table 4 (test 2) shows the number of firms in each category in each year. Results are presented in Table 5. The null hypothesis of spanning is rejected for firms with foreign sales less than 5 per cent, showing that these firms yield consistent diversification benefits. The market value weighted test shows an 11 per cent improvement in the Sharpe ratio for the whole period and a 4.45 per cent increase for the pre-crisis sub-period, but the crisis and post-crisis periods show no benefits. The benefits using equally weighted portfolios are even larger, with 39.71 per cent increase for the whole period, 8.96 per cent for the pre-crisis period, 34.49 per cent for the post-crisis period and 16.53 per cent for the crisis period. However, for the remaining two categories, approximately half of the Wald tests fail to reject spanning. We find no benefits from investing in firms with 5–15 per cent foreign sales and only marginal benefits from investing in firms with more than 15 per cent foreign sales. We conclude that the benefits from investing in MNCs in China are marginal and firms with greater levels of multi-nationality do not provide greater benefits.

Table 3.
MVS Results using ABHK Classifications

Time period

Full period Pre-crisis Crisis Post-crisis

Panel A: All MNCs

Market value weighted α = 0 and β = 1 0.0000 0.0118 0.0000 0.0000

α = 0 0.6712 0.6441 0.0679 0.1722

β = 1 0.0000 0.0030 0.0000 0.0000

SR Change (%) 0.00% 0.57% 0.00% 0.00%

Equally weighted α = 0 & β = 1 0.1309 0.2154 0.3671 0.5616

α = 0 0.5452 0.2116 0.3168 0.7246

β = 1 0.0497 0.1442 0.2947 0.3200

SR Change (%) Fail to reject Null

Panel B: Market Value Weighted Portfolios

Extended Set Benchmark SR 0.0633 0.1794 −0.0349 −0.0414

T2 foreign sales less than 5% α = 0 & β = 1 0.0000 0.0000 0.0000 0.0000

α = 0 0.4814 0.9309 0.3109 0.4044

β = 1 0.0000 0.0000 0.0000 0.0000

SR Change (%) 0.03% 0.00% 0.00% 0.00%

T2 foreign sales between 5% and 15% α = 0 & β = 1 0.0000 0.0000 0.0000 0.0000

α = 0 0.4794 0.4779 0.1535 0.2906

β = 1 0.0000 0.0000 0.0000 0.0000

SR Change (%) 0.00% 2.18% 0.00% 0.00%

T2 foreign sales above 15% α = 0 & β = 1 0.0000 0.0000 0.0002 0.0000

α = 0 0.0277 0.7807 0.0668 0.1007

β = 1 0.0000 0.0000 0.0001 0.0000

SR Change (%) 0.00% 0.00% 0.00% 0.00%

T3,T4,T5,G α = 0 & β = 1 0.0000 0.0000 0.0349 0.0118

α = 0 0.8899 0.1214 0.2193 0.4714

β = 1 0.0000 0.0000 0.0200 0.0036

SR Change (%) 0.00% 11.49% 0.00% 0.00%

Panel C: Equally Weighted Portfolios

Benchmark SR 0.0309 0.1371 −0.0467 −0.0658

T2 foreign sales less than 5% α = 0 & β = 1 0.0000 0.0000 0.0000 0.0000

α = 0 0.7823 0.5311 0.4125 0.9211

β = 1 0.0000 0.0000 0.0000 0.0000

SR Change (%) 2.88% 2.95% 0.00% 17.37%

T2 foreign sales between 5% and 15% α = 0 & β = 1 0.0000 0.0002 0.0000 0.0000

α = 0 0.9829 0.8806 0.5614 0.7137

β = 1 0.0000 0.0001 0.0000 0.0000

SR Change (%) 0.00% 0.00% 0.00% 36.40%

T2 foreign sales above 15% α = 0 & β = 1 0.0000 0.0000 0.0001 0.0000

α = 0 0.9727 0.1683 0.5293 0.6486

β = 1 0.0000 0.0000 0.0000 0.0000

SR Change (%) 0.00% 14.74% 0.00% 0.00%

T3,T4,T5,G α = 0 & β = 1 0.0000 0.0001 0.0556 0.0666

α = 0 0.6318 0.4577 0.2109 0.5216

β = 1 0.0000 0.0000 0.0348 0.0233

SR Change (%) 0.00% 4.38% 0.00% 0.00%

		Time period
Panel A: All MNCs
Market value weighted	α = 0 and β = 1	0.0000	0.0118	0.0000	0.0000
α = 0	0.6712	0.6441	0.0679	0.1722
β = 1	0.0000	0.0030	0.0000	0.0000
SR Change (%)	0.00%	0.57%	0.00%	0.00%
Equally weighted	α = 0 & β = 1	0.1309	0.2154	0.3671	0.5616
α = 0	0.5452	0.2116	0.3168	0.7246
β = 1	0.0497	0.1442	0.2947	0.3200
SR Change (%)	Fail to reject Null
Panel B: Market Value Weighted Portfolios
Extended Set	Benchmark SR	0.0633	0.1794	−0.0349	−0.0414
T2 foreign sales less than 5%	α = 0 & β = 1	0.0000	0.0000	0.0000	0.0000
α = 0	0.4814	0.9309	0.3109	0.4044
β = 1	0.0000	0.0000	0.0000	0.0000
SR Change (%)	0.03%	0.00%	0.00%	0.00%
T2 foreign sales between 5% and 15%	α = 0 & β = 1	0.0000	0.0000	0.0000	0.0000
α = 0	0.4794	0.4779	0.1535	0.2906
β = 1	0.0000	0.0000	0.0000	0.0000
SR Change (%)	0.00%	2.18%	0.00%	0.00%
T2 foreign sales above 15%	α = 0 & β = 1	0.0000	0.0000	0.0002	0.0000
α = 0	0.0277	0.7807	0.0668	0.1007
β = 1	0.0000	0.0000	0.0001	0.0000
SR Change (%)	0.00%	0.00%	0.00%	0.00%
T3,T4,T5,G	α = 0 & β = 1	0.0000	0.0000	0.0349	0.0118
α = 0	0.8899	0.1214	0.2193	0.4714
β = 1	0.0000	0.0000	0.0200	0.0036
SR Change (%)	0.00%	11.49%	0.00%	0.00%
Panel C: Equally Weighted Portfolios
	Benchmark SR	0.0309	0.1371	−0.0467	−0.0658
T2 foreign sales less than 5%	α = 0 & β = 1	0.0000	0.0000	0.0000	0.0000
α = 0	0.7823	0.5311	0.4125	0.9211
β = 1	0.0000	0.0000	0.0000	0.0000
SR Change (%)	2.88%	2.95%	0.00%	17.37%
T2 foreign sales between 5% and 15%	α = 0 & β = 1	0.0000	0.0002	0.0000	0.0000
α = 0	0.9829	0.8806	0.5614	0.7137
β = 1	0.0000	0.0001	0.0000	0.0000
SR Change (%)	0.00%	0.00%	0.00%	36.40%
T2 foreign sales above 15%	α = 0 & β = 1	0.0000	0.0000	0.0001	0.0000
α = 0	0.9727	0.1683	0.5293	0.6486
β = 1	0.0000	0.0000	0.0000	0.0000
SR Change (%)	0.00%	14.74%	0.00%	0.00%
T3,T4,T5,G	α = 0 & β = 1	0.0000	0.0001	0.0556	0.0666
	α = 0	0.6318	0.4577	0.2109	0.5216
β = 1	0.0000	0.0000	0.0348	0.0233
SR Change (%)	0.00%	4.38%	0.00%	0.00%

Source: Authors’ own calculations.

Notes: This table shows the results from our MVS tests. Domestic firms are used as the benchmark portfolio. Firms are classified based on sales data and short sales are not allowed. Panel A shows the results of MVS tests between Chinese domestic firms and all MNCs. Panel B details results of MVS tests between domestic firms and firms of different multinationality levels using market value weighted returns. Panel C shows results of MVS tests between domestic firms and firms of different multinationality levels using equally weighted returns. The p-values are from the Wald test of the joint hypothesis that α = 0 and β = 1, and the results from the step-down Wald tests. We also show the increase in the Sharpe ratios of the optimal portfolios.

Our results show that Chinese investors can benefit from international portfolio diversification. However, Chinese MNCs do not yield significant diversification benefits, but firms with lower levels of multinationality provide more consistent diversification benefits. The indirect diversification benefits attained do not increase as firm-level multinationality increases. Our sub-period analysis shows that no benefits exist during the crisis period and benefits tend to be higher in the post-crisis period compared to the pre-crisis period.

Table 4.

Number of Firms in Each Category

Year	Benchmark	Test 1				Test 2
Year	D	T2 <=5%	5%<T2<=15%	T2>15%	T3,T4,T5,G	Below 5%	5% to 15%	Greater than 15%
2002	30	5	5	6	3	5	6	8
2003	33	5	10	8	3	5	11	10
2004	39	7	9	9	4	7	10	12
2005	60	9	14	12	4	9	15	15
2006	63	10	13	17	4	12	13	21
2007	60	16	13	16	8	17	13	24
2008	53	18	18	20	7	19	19	27
2009	52	25	21	16	8	26	26	21
2010	52	27	16	19	8	29	19	26
2011	51	26	16	17	11	29	20	25
2012	52	32	26	34	9	36	28	42
AVG	49.55	16.36	14.64	15.82	6.27	17.64	16.36	18.90

Source: Authors’ own calculations.

Notes: This table shows the number of firms in each category. For test 1, regional MNCs are not included as the number is limited, and a maximum of 10 firms are chosen for each MVS test. Test 2 shows foreign sales groups for MVS tests. All firms listed in the table are used in MVS tests.

Table 5.

MVS Results using Foreign Sales Categories

		Full period	Pre-crisis	Crisis	Post-crisis
Panel A: Market Value Weighted Portfolios
Extended set	Benchmark SR	0.0861	0.2403	−0.0542	−0.0577
FS < 5%	α = 0 & β = 1	0.0000	0.0506	0.0004	0.0000
	α = 0	0.2940	0.2172	0.7573	0.5329
	β = 1	0.0000	0.0189	0.0001	0.0000
	SR Change (%)	11.13%	4.45%	0.00%	0.00%
FS > 5% FS < 15%	α = 0 & β = 1	0.0004	0.3923	0.0012	0.4108
FS > 5% FS < 15%	α = 0	0.4446	0.8710	0.0937	0.4676
	β = 1	0.0002	0.1752	0.0007	0.2506
	SR Change (%)	0.00%	Fail to reject Null	0.00%	Fail to reject Null
FS > 15%	α = 0 & β = 1	0.0000	0.0121	0.0026	0.0000
	α = 0	0.4315	0.6932	0.0738	0.0592
	β = 1	0.0000	0.0032	0.0022	0.0000
	SR Change (%)	0.00%	0.40%	0.00%	0.00%
Panel B: Equally Weighted Portfolios
	Benchmark SR	0.0618	0.1955	−0.0548	−0.0644
FS <5%	α = 0 & β = 1	0.0020	0.2107	0.0904	0.0002
	α = 0	0.0824	0.1509	0.8078	0.3677
	β = 1	0.0014	0.0148	0.0308	0.0001
	SR Change (%)	39.71%	8.96%	16.53%	34.49%
FS > 5% FS < 15%	α = 0 & β = 1	0.7568	0.0671	0.0263	0.8758
	α = 0	0.4559	0.1888	0.6347	0.6926
	β = 1	0.9798	0.0319	0.0076	0.7273
	SR Change (%)	Fail to reject Null	0.00%	0.00%	Fail to reject Null
FS > 15%	α = 0 & β = 1	0.8589	0.0013	0.5056	0.0851
	α = 0	0.8685	0.0310	0.0846	0.8665
	β = 1	0.5927	0.0009	0.1847	0.2526
	SR Change (%)	Fail to reject Null	17.67%	Fail to reject Null	Fail to reject Null

Source: Authors’ own calculations.

Notes: This table shows the results from our MVS tests. Domestic firms are used as the benchmark portfolio. Firms are classified based on sales data and short sales are not allowed. Panel A shows the results of MVS tests using market value weighted portfolios and Panel B shows results using equally weighted portfolios. The p-values are from the Wald test of the joint hypothesis that α = 0 and β = 1, and the results from the step-down Wald tests. We also show the increase in the Sharpe ratios of the optimal portfolios.

5. Conclusion

We provide a unique longitudinal analysis of firm-level multinationality for Chinese firms between 2002 and 2012, using the Aggarwal et al. (2011) classification system and the percentage of foreign sales. Results show that most Chinese MNCs are categorised as T2 and have relatively low figures for foreign sales—the level and breadth of firm-level multinationality is low for firms in China. While the majority of sales are domestic, there has been some increase in foreign sales over the sample period. Europe and the USA are China’s most important exporting regions but exports to both regions reduced significantly since 2007. Our results suggest that the increase in foreign sales attributes more to the growth in number of MNCs rather than further advancement of existing MNCs.

We investigate the benefits from investing in Chinese MNCs using MVS and Sharpe ratio tests. We conclude that no extra benefits can be created by investing in MNCs. This means on average MNC shares do not perform better than domestic shares in China. Although we conclude that Chinese MNCs, in general, do not provide significant diversification benefits, counter-intuitively, our results show that firms with lower levels of multinationality tend to provide some benefits. We also conclude that the benefits from indirect portfolio diversification do not exist during the crisis period. We test the robustness of our results using two measures of multinationality, both market value and equally weighted portfolios and across three subsample time periods.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Footnotes

Notes

References

Aggarwal

, Berrill

, Hutson

, & Kearney

(2011). What is a multi-national corporation? Classifying the degree of firm-level multinationality. International Business Review, 20(5), 557–577.

Agmon

, & Lessard

(1977). Investor recognition of corporate international diversification. The Journal of Finance, 32(4), 1049–1055.

Aharoni

(2006). Book review: Alan M Rugman, the regional multinationals. MNEs and global strategic management. International Business Review, 15(4), 439–446.

Ahearne

, Griever

, & Warnock

(2004). Information costs and home bias: An analysis of US holdings of foreign equities. Journal of International Economics, 62(2), 313–336.

Antoniou

, Olusi

, & Paudyal

(2010). Equity home bias: A sub-optimal choice for UK investors. European Financial Management, 16(3), 449–479.

Asmussen

(2008). Local, regional, or global; quantifying MNE geographic scope. Journal of International Business Studies, 40(7), 1192–1205.

Bai

, & Perron

(1998). Estimating and testing linear models with multiple structural changes. Econometrica, 66(1), 47–78.

Berrill

(2015). Are the world’s largest firms regional or global? Thunderbird International Business Review, 57(2), 87–101.

Berrill

, & Kearney

(2010). Firm-level internationalisation and the home bias puzzle. Journal of Economics and Business, 62(4), 235–256.

10.

Brewer

(1981). Investor benefits from corporate international diversification. Journal of Financial and Quantitative Analysis, 16(1), 113–126.

11.

Brouthers

, & Hennart

J. F.

(2007). Boundaries of the firm: Insights from international entry mode research. Journal of Management, 33(3), 395–425.

12.

Buckley

, & Chapman

(1997). The use of native categories in management research. British Journal of Management, 8(4), 283–299.

13.

Cai

, & Warnock

F. E.

(2004). International diversification at home and abroad (International Finance Discussion Papers No. 793). Washington, DC: Board of Governors of the Federal Reserve System.

14.

Canabal

, & White

(2008). Entry mode research: Past and future. International Business Review, 17(3), 267–284.

15.

Casillas

, & Acedo

(2013). Speed in the internationalisation process of the firm. International Journal of Management Reviews, 15(1), 15–29.

16.

Chow

(1960). Tests of equality between sets of coefficients in two linear regressions. Econometrica, 28(3), 591–605.

17.

Collinson

, & Rugman

(2008). The regional nature of Japanese multinational business. Journal of International Business Studies, 39(2), 215–230.

18.

Contractor

(2007). Is international business good for companies? The evolutionary or multi-stage theory of internationalisation vs. the transaction cost perspective. Management International Review, 47(3), 453–475.

19.

De Roon

, & Nijman

(2001). Testing for mean variance spanning: A survey. Journal of Empirical Finance, 8(2), 111–156.

20.

Doremus

, Keller

, Pauly

, & Reich

(1998). The myth of the global corporation. Princeton, NJ: Princeton University Press.

21.

Driessen

, & Laeven

(2007). International portfolio diversification benefits: Cross-country evidence from a local perspective. Journal of Banking and Finance, 31(6), 1693–1712.

22.

Dunning

, Fujita

, & Yakova

(2007). Some macro-data on the regionalisation/globalisation debate: A comment on the Rugman/Verbeke analysis. Journal of International Business Studies, 38(1), 177–199.

23.

Errunza

, Hogan

, & Hung

(1999). Can the gains from international diversification be achieved without trading abroad? The Journal of Finance, 54(6), 2075–2107.

24.

Fatemi

(1984). Shareholder benefits from corporate international diversification. The Journal of Finance, 39(5), 1325–1344.

25.

Fillis

(2001). Small firm internationalisation: An investigative survey and future research directions. Management Decision, 39(9), 767–783.

26.

Ghemawat

(2001). Distance still matters: The hard reality of global expansion. Harvard Business Review, 79(8), 137–147.

27.

Ghemawat

(2003). Semiglobalisation and international business strategy. Journal of International Business Studies, 34(1), 138–152.

28.

Glaum

, & Oesterle

(2007). 40 years of research on internationalisation and ﬁrm performance: More questions than answers? Management International Review, 47(3), 307–317.

29.

Govindarajan

, & Gupta

(2008). The quest for global dominance: Transforming global presence into global comparative advantage (2nd ed.). San Francisco, CA: Jossey-Bass.

30.

Hennart

J. F.

(2007). The theoretical rationale for a multinationality/performance relationship. Management International Review, 47(3), 423–453.

31.

Huberman

, & Kandel

(1987). Mean-variance spanning. The Journal of Finance, 42(4), 873–888.

32.

Hughes

, Logue

, & Sweeney

(1975). Corporate international diversification and market assigned measures of risk and diversification. Journal of Financial and Quantitative Analysis, 10(4), 627–637.

33.

Jacquillat

, & Solnik

(1978). Multinationals are poor tools for diversification. Journal of Portfolio Management, 4(2), 8–12.

34.

Johanson

, & Vahlne

(1990). The mechanism of internationalization. International Marketing Review, 7(4), 11–24.

35.

Kan

, & Zhou

(2012). Tests of mean-variance spanning. Annals of Economics and Finance, 13(1), 139–187.

36.

Kim

, & Lyn

(1990). FDI theories and the performance of foreign multinationals operating in the U.S. Journal of International Business Studies, 21(1), 41–54.

37.

Leonidou

, & Katsikeas

(1996). The export development process: An integrative review of empirical models. Journal of International Business Studies, 27(3), 517–551.

38.

Levy

, & Sarnat

(1970). International diversification of investment portfolios. American Economic Review, 60(4), 668–675.

39.

Logue

(1982). An experiment in international diversification. Journal of Portfolio Management, 9(1), 22–27.

40.

Mathur

, Singh

, & Gleason

(2001). The evidence from Canadian firms on multinational diversification and performance. The Quarterly Review of Economics and Finance, 41(4), 561–578.

41.

Michel

, & Shaked

(1986). Multinational corporations versus domestic corporations: Financial performance and characteristics. Journal of International Business Studies, 17(3), 89–100.

42.

Mikhail

, & Shawky

(1979). Investment performance of US based multinational corporations. Journal of International Business Studies, 10(1), 53–66.

43.

Ohmae

(1985). Triad power: The coming shape of global competition. New York, NY: The Free Press.

44.

Osegowitsch

, & Sammartino

(2007). Exploring trends in regionalisation. In Rugman

A.M.

(Ed.), Research in global strategic management (Vol. 13, Regional aspects of multinationality and performance, pp. 45–64). Amsterdam: Elsevier.

45.

Osegowitsch

, & Sammartino

(2008). Reassessing (home-) regionalization. Journal of International Business Studies, 39(2), 184–196.

46.

Rowland

, & Tesar

(2004). Multinationals and the gains from international diversification. Review of Economic Dynamics, 7(4), 789–826.

47.

Rugman

(2000). The end of globalization. London, UK: Random House.

48.

Rugman

(2003). Regional strategy and the demise of globalization. Journal of International Management, 9(4), 409–417.

49.

Rugman

(Ed.) (2005). The regional multinationals: MNEs and ‘global’ strategic management. Cambridge, UK: Cambridge University Press.

50.

Rugman

, & Brain

(2003). Multinational enterprises are regional, not global. Multinational Business Review, 11(1), 3–12.

51.

Rugman

, & Girod

(2003). Retail multinationals and globalization. European Management Journal, 21(1), 24–37.

52.

Rugman

, & Hodgetts

(2001). The end of global strategy. European Management Journal, 19(4), 333–343.

53.

Rugman

, & Oh

(2010). Does the regional nature of multinationals affect the multinationality and performance relationship? International Business Review, 19(5), 479–488.

54.

Rugman

, & Verbeke

(2003). Regional transnationals and triad strategy. Transnational Corporation, 13(3), 1–20.

55.

Rugman

, & Verbeke

(2004). A perspective on regional and global strategies of multinational enterprises. Journal of International Business Studies, 35(1), 3–18.

56.

Rugman

, & Verbeke

(2007). Liabilities of regional foreignness and the use of firm-level versus country-level data: A response to Dunning et al. (2007). Journal of International Business Studies, 38(1), 200–205.

57.

Rugman

, & Verbeke

(2008). The theory and practice of regional strategy: A response to Osegowitsch and Sammartino. Journal of International Business Studies, 39(2), 326–332.

58.

Salehizadeh

(2003). US multinationals and the home bias puzzle: An empirical analysis. Global Finance Journal, 14(3), 303–318.

59.

Senchack

, & Beedles

(1980). Is indirect international diversification desirable? Journal of Portfolio Management, 6(2), 49–57.

60.

Solnik

(1974). Why not diversify international rather than domestically? Financial Analysts Journal, 30(4), 48–54.

61.

Song

(2004). How to identify China capital market’s risk free rate. Journal of Guangxi Financial Research, 9(1), 33–35.

62.

Suh

(2005). Home bias among institutional investors: A study of the quarterly portfolio poll. Journal of the Japanese and International Economies, 19(1), 72–95.

63.

Welch

, & Luostarinen

(1988). Internationalisation: Evolution of a concept. Journal of General Management, 14(2), 34–55.

64.

Wright

, & McCarthy

(2002). Does purchasing stock in Australian multi-national corporations create international portfolio diversification? Multi-national Business Review, 10(1), 79–83.

65.

Xing

(2003). Internationalisation of China’s firms: The current situation, problems and coping strategies. Reform of Economy System, 1(1), 30–33.

66.

Yip

(2002). Total global strategy II. Upper Saddle River, NJ: Prentice Hall.