Đầu tư ở nước nhà hay đầu tư ra nước ngoài? Quan hệ giữa đầu tư trực tiếp ra nước ngoài và hiệu quả hoạt động của các công ty Đài Loan

Nội dung text: Đầu tư ở nước nhà hay đầu tư ra nước ngoài? Quan hệ giữa đầu tư trực tiếp ra nước ngoài và hiệu quả hoạt động của các công ty Đài Loan

1. HOME & AWAY ? THE RELATIONSHIP BETWEEN OUTWARD INVESTMENT AND PERFORMANCE FOR TAIWANESE COMPANIES ĐẦU TƯ Ở NƯỚC NHÀ HAY ĐẦU TƯ RA NƯỚC NGOÀI? QUAN HỆ GIỮA ĐẦU TƯ TRỰC TIẾP RA NƯỚC NGOÀI VÀ HIỆU QUẢ HOẠT ĐỘNG CỦA CÁC CÔNG TY ĐÀI LOAN Pan, Weihwa Yang, Cheng-Hung National Yunlin University of Science and Technology Abstract This study distinguishes outward investment regions into those in approximative and non-approximative regions investment by the perspective of “Liability of Foreignness” (LOF), which examines the influence between different outward investment regions and performance, also, investigates the different results by industry. Internationalization and product diversification are served as moderators to examine their influence to outward investment and performance. Finally, this study concludes the relationship of different configuration of both regions and moderators performance. This study applies GLS and MLM to test the sample of 6783 observations from 544 listed firms in 18 different industries dated from 2002 to 2015. The empirical results show the inverted U-shaped relationship between the levels of approximative region investment and performance, and J-shaped relationship between the levels of non-approximative region investment and performance. Better performance can be achieved for firms in higher degree of internationalization. Product diversification suggest a discount effect. The firms in lower degree of product diversification are outperformed than those in higher degree. There is better performance at the configuration of high degree of internationalization and low or high level of non-approximative region investment, as well as, at the configuration of low degree of product diversification and medium level of approximative region investment. As for industry, semiconductor industry is the best performer in outward investment, while optoelectronic industry is the worst one. Keywords: Outward investment, Liability of Foreignness, Internationalization, Product diversification Tóm tắt Nghiên cứu tìm hiểu về đầu tư trực tiếp ra nước ngoài vào các vùng đầu tư lân cận và không lân cận theo tiếp cận từ trách nhiệm quốc tế (LOF), xem xét ảnh hưởng giữa đầu tư trực tiếp ra nước ngoài và hiệu quả đầu tư, nghiên cứu các kết quả đầu tư khác nhau giữa các ngành công nghiệp khác nhau. Quốc tế hoá và đa dạng hoá sản phẩm là các nhân tố được xem xét để tìm hiểu mức độ ảnh hưởng của chúng đối với hoạt động đầu tư ra nước ngoài và hiệu quả đầu tư. Cuối cùng, nghiên cứu đưa ra kết luận về mối quan hệ giữa các nhân tố đối với vùng đầu tư và hiệu quả đầu tư. Nghiên cứu áp dụng phương pháp GLS và MLM để kiểm định mẫu gồm 6783 quan sát từ 544 công ty niêm yết thuộc 18 ngành công nghiệp khác nhau trong khoảng thời gian 148
2. từ năm 2002 đến 2015. Kết quả thực nghiệm cho thấy có mối quan hệ hình chữ U ngược giữa mức độ đầu tư ở vùng lân cận và hiệu quả đầu tư, có quan hệ hình chữ J giữa mức độ đầu tư ở vùng phi lân cận và hiệu quả đầu tư. Hiệu quả hoạt động ở các doanh nghiệp có mức độ quốc tế hoá cao hơn là tốt hơn. Đa dạng hoá sản phẩm có quan hệ tỉ lệ nghịch. Các doanh nghiệp có tỷ lệ đa dạng hoá sản phẩm thấp hơn lại có hoạt động hiệu quả hơn so với những doanh nghiệp có tỉ lệ đa dạng hoá cao hơn. Hiệu quả hoạt động là tốt hơn ở các doanh nghiệp có mức độ quốc tế hoá cao và mức độ phi lân cận thấp hoặc cao, cũng như vậy hiệu quả hoạt động là tốt hơn ở doanh nghiệp có mức độ đa dạng hoá sản phẩm thấp và mức độ vùng lân cận trung bình. Đối với các ngành, ngành công nghiệp bán dẫn có hiệu quả cao nhất trong đầu tư ra nước ngoài, trong khi ngành sản xuất quang điện tử có hiệu quả thấp nhất. Từ khoá: Đầu tư ra nước ngoài, trách nhiệm quốc tế, quốc tế hoá, đa dạng hoá sản phẩm 1. Introduction Since 1980s, Taiwanese firms started to conduct outward investment as a result of appreciation of NTD, increasing costs of production and rising concern of environmental pollution. According to statistics from Ministry of Economic Affairs, the outward investment of Taiwan had reached 21.1 billion USD in 2015, yet a half of investment flowed to the region of mainland China, Hong Kong and Macao, a geographical and cultural approximative region. The proportion of investment to China was 49% in 2015 and the average proportion was 60.6% from 2011 to 2015. According to the above information, the outward investment of Taiwan highly concentrates to mainland China. However, there is a rising concern about whether the over-concentrating outward investment would cause the negative influence to firm’s performance or not? Therefore, to test this concern has become the main theme of this dissertation. Over the past decades, there were different results of the relationship between outward investment and performance, some of scholars indicated that the relationship between them was inverted U-shaped, but some was U-shaped. One possible reason to cause different result might be that previous researches can’t examine the different effects caused by different contexts. Hence, this study contributes to literature by differentiating outward investment region according to the level of liability of foreignness (Hymer, 1960 & 1976). In addition, level of LOF could measure by CAGE distance framework (Ghemawat, 2001) which including cultural, administrative, geographic and economic distances. The region in lower LOF level will be foreign approximative regions, such as mainland China, Hong Kong, Macao which are in same language, frequent interaction of both government, near physical distance with same time zone and integrated economy, and others will be non- approximative regions. Although Singapore is also a Chinese region, the industrial structures are different from the industries in this research. Finding out the relationship between outward investment and performance by different contexts is one of the motivations of this study. 149
3. Furthermore, outward investment is a means of internationalization. While the production costs increased or market saturated in home countries, firms can explore the possibility of new markets or new resources as a way to expand internationally. Consequently, firms would apply the resources they get internationally for further expansion again (Deresky, 1994). In other words, firms are internationalized by outward investment, and conduct ongoing investment to other places. Therefore, examining how internationalization influence the relationship of outward investment and performance is another motivation of this study. Moreover, there is a possible relationship between outward investment and product diversification. For firms, outward investment can usually obtain new resources and new markets which may be instrumental for conducting the product diversification. In other words, outward investment be a way to obtain the required new resources, or tap into new markets for newly-diversified business. Thus, it is reasonable to assume the moderating effect of product diversification when discussing the different effect between outward investment and performance. How different the level of product diversification affecting the relationship between outward investment and performance is another motivation of this study. Finally, previous researches (Gomes & Ramaswamy, 1999; Kundu, Hsu and Li, 2003) didn’t examine the relationship between outward investment and performance by single-out the industrial effect. Obviously, there would be different influence because of different industrial competition from different host countries. Consequently, examining different performance from outward investment of each industry is final motivation of this study. In sum, outward investment is an important strategic means of enterprise growth. While the profit margin gradually reduced as the industrial development, outward investment could give the new opportunity to firms. Hence, there is an importance for the research of outward investment in international business studies. 2. Literature Review & Hypothesis Development 2.1.Outward Investment and Performance In order to survive, enterprises have to find an exit for themselves once profit margin has been squeezed in home countries. Outward investment is a strategy for firms that expands operations to foreign countries through greenfield investment, merger, acquisition, joint venture or expansion of existing foreign facilities by cash and stocks. The motivations to conduct outward investment are concluded as: (1) market-seeking, (2) resource-seeking, (3) efficiency-seeking, (4) strategic asset-seeking (Dunning, 1993 &2000). Outward investment is a natural motion for firms to seek better business opportunities which could be available in foreign countries. However, according to previous researches, the relationship between outward investment and performance is inconsistent. Gomes and Ramaswamy (1999) considered that there is inverted U-shaped relationship between them. They indicated the sales and 150
4. profits would be increased because of the new markets or lower costs of production and resources had been explored, until the development reach to specific level, firms not only involved the favorable markets, but also confronted with the competition from competitors or emerging firms from host countries, and undertaken the instability of host countries causing performance decreased. Hence, the inverted U-shape came out. Conversely, Contractor, Kundu, Hsu and Li (2003) stated the relationship of them is U-shaped. They demonstrated that there were some extra governance and cooperated costs, while the firms entered to new region. In other words, there was a “tuition” for firms to pay at the initial period of entry new place. Then, through the development, firms would be familiar with the operation systems and increase the performance. Thus, the U-shape came out. Liability of foreignness (LOF) demonstrated that there are some costs containing (1) limitation of local knowledge and information, such as legal and nationalism of host countries (2) coordination costs of host stakeholders, such as information asymmetry from benefits distributing (3) costs from home countries, such as the restriction of technique transferring (4) firm-specific costs because of unfamiliarity and the lack of rooting (Zaheer, 1995). According to Gaur, Kumar and Sarathy (2011), there are two origin causing LOF: (1) environment-derived which including different governments, institutions, cultures, the nature and structure of industries (Nachum, 2003; Zaheer, 1995) (2) firm- based which derived from firm-specific characteristics including ownership structure, firm- specific resources, learning, and network of business groups (Johanson & Vahlne, 2009; Petersen & Pedersen, 2002). The perspective of LOF bridges the gap from the inconsistent result of the relationship between outward investment and performance. While the firms invest to the region in lower LOF, there are lower extra governance and coordinated costs, but new markets and new resources, causing profits larger than costs, until the specific level of development achieve. Thus, the inverted U-shaped relationship between outward investment and performance take place. On the other hand, while the firms invest to region in higher LOF, there is “tuition” they need to pay causing costs larger than profits before getting better. Hence, the U-shape occurs. In sum, there might be different results while firms invest to different context region. CAGE distance framework (Ghemawat, 2001) which identifies cultural, administrative, geographic and economic differences would be a way to measure LOF. Cultural distance includes languages, ethnicities and religions; administrative distance includes regional trading, currency and political interaction; geographic distance includes physical distance, time zone and physical environments; economic distance includes rich/poor differences and economic interaction. In this study, we differentiate outward investment into approximative and non-approximative region investment by the perspective of liability of foreignness in order to examining the different relationship of outward investment from different context and performance. The approximative region for Taiwanese firms outside Taiwan is defined as mainland China, Hong Kong, Macao which are in Chinese context with same languages and ethnicities, frequently political interaction, near physical distance, same time zone, highly economic interaction, and other countries are categorized to non-approximative region. Furthermore, although Singapore is a country 151
5. in Chinese context, the industry structures differ from the industries in this research. Therefore, through the reason above, the hypothesis are: Hypothesis 1A: The relationship between approximative region investment and performance is significant inverted U-shape. Hypothesis 1B: The relationship between non-approximative region investment and performance is significant U-shape. 2.2. Effect of Internationalization to Outward Investment and Performance The relationship between internationalization and performance exists inconsistent argument. Some of scholars reported that the relationship between internationalization and performance is positive (Vernon, 1971; Errunza & Senbet, 1984; Grant, 1987), some are negative (Siddharthan & Lall, 1982; Geringer, Tallman & Olsen, 2000), even some are U- shaped (Geringer, Beamish & daCostsa, 1989), inverted U-shaped ((Lu & Beamish, 2001) and S-shaped (Farok J. Contractor et al., 2003; Thomas & Eden, 2004). Outward investment is one of the means of internationalization. In general, firms would conduct export, license and franchise to expand new oversea markets. However, some host countries would set some barriers such as tariff barriers and ownership restriction, etc. to protect local industries. In order to entering those countries, firms would conduct economic investment such as joint venture by abiding the regulation of host countries. On the other hand, to obtaining specific resources from a particular country, firms would also conduct outward investment activities to set up facilities or joint venture. In sum, firms used outward investment to avoid some restriction or to obtain the resources they need. Moreover, firms might continually invest to next place for new markets, new resources and profits. Thus, the relationship between outward investment and internationalization is significant, because firms use outward investment to internationalize then expand to other places by outward investment. Additionally, the firms with different degree of internationalization would undertake different risks and costs. The firms with high degree internationalization would inevitably bear huge operating and managerial costs. In other words, internationalization would increase the complexity for firms. On the other hand, the firms with low degree internationalization would cost less operating and managerial costs, but couple with transaction cost and learning expenses. While the firms conduct outward investment, the performance might be different by different degree of internationalization because risks and costs they undertake would be different. To sum up, this study would examine the effect of internationalization to the relationship between outward investment and performance. The hypothesis would be: Hypothesis 2: The degree of internationalization would moderate the relationship between outward investment and performance. 2.3. Effect of Product Diversification to Outward Investment and Performance There are inconsistent argument about the relationship between product diversification and performance. Some of scholars indicated that the relationship between them is positive, such as Chandler (1977), Teece (1982) and Williamson (1975). They 152
6. discovered that the performance of the firms which with product diversification was better than the firms which without product diversification. Others considered that the relationship between them is negative, such as Lang and Stulz (1994), Berger and Ofek (1995) and Servaes (1996). They indicated that there would be more costs and risks while conduct product diversification causing performance decreasing. Furthermore, other scholars outcropped the relationship between product diversification and performance is inverted U-shaped (Grant, Jammine & Thomas, 1988; Lubatkin & Chatterjee, 1994; Palich et al., 2000). They considered that firms would conduct relative product diversification while the initial period causing performance increasing until a specific level which firms must to conduct non-relative product diversification causing performance starting to decrease. In other words, firms would develop relative product diversification while beginning causing performance increasing by the exploration of new target markets. However, while a specific level of development, the relative product markets would be saturated leading firms to have to conduct non-relative product diversification causing performance decreasing because of extra developing costs by non-experience. Because of entering new scope by product diversification, it’s necessary for firms to seek the technique and resources of this new field. Outward investment would be a solution of this question. While firms conduct outward investment, some kind of resources or technique endowed in this countries could be acquired. Moreover, because of market saturation of diversified product, outward investment would also be a way to explore new markets. Therefore, there might be a relationship between outward investment and product diversification because firms would use outward investment to seek the resources for product diversification and to seek the markets for the product they develop. In sum, the firms with different degree of product diversification might undertake different risks and costs. Product diversification would also increase the complexity for firms causing performance might be different, while firms conduct outward investment with different degree of product diversification. Thus, this study would examine the effect of product diversification to the relationship between outward investment and performance. The hypothesis would be: Hypothesis 3: The degree of product diversification would moderate the relationship between outward investment and performance. 3. Methodology 3.1. Research Structure This study aims to examine the relationship between different regions of outward investment and it’s implication to performance. Therefore, outward investment would be distinguished to approximative and non-approximative region investment which are measured by the proportion to total investment. Moreover, internationalization and product diversification would be considered as moderators which are probably influence the relationship between outward investment and performance. The performance would be presented by ROA. The research structure is shown below Figure 3-1: 153
7. Figure 3-1 Research Structure Figure 3-2 Structure of Multilevel Model 3.2. Data Collection & Description and Research Method This study acquires the data from Taiwan Economic Journal Co., Ltd (TEJ Database) which is frequently used in Taiwan. The data in this study are panel data from 18 manufacturing industries which are categorized by Taiwan Stock Exchange including 544 listed firms dated from 2002 to 2015, totally 6783 firm-year observations. There are two statistical methods applied in our research, which are panel regression of generalized least square (GLS) method and multilevel model (MLM). GLS which is the most commonly used in linear regression model is a method which estimates unknown parameters by the sum of squared vertical distance between the observations and the predicted regression line. In addition, multilevel linear model (MLM) is fit to investigate the relationship between independent variable and dependent variable at each clusters. In our study, data are nested for 3 level, particularly, industries data are nested in level 1, the firms of each industries are nested in level 2, and the observations of years from firms are nested in level 3. MLM is different from GLS because it could examine the relationship between independent variables and dependent variables from each clusters at the same time. Thus, it’s the fittest for us to find out the relationship between outward investment and performance by each industries. The multilevel structure is shown above Figure 3-2. 3.3.Measurements (1) Dependent Variable Performance Return on assets (ROA) is the one of the widely-used indicators for measuring firm performance (Daniels & Bracker, 1989; Haar, 1989; Ramaswamy, 1995; Gomes & 154
8. Ramaswamy, 1999). The higher ROA demonstrates better performance, indicating higher efficiency of firms to generate profits by employing assets. It’s calculated by dividing net income of firm by total assets. (2)Independent Variable Outward Investment This study would differentiate outward investment into two mutually exclusive constructs of approximative and non-approximative region investment. For Taiwan, China, Hong Kong and Macao are the places we consider as approximative regions by level of LOF measuring from CAGE frameworks. Moreover, lots of Taiwanese firms invest in other third-party countries such as BVI, Cayman Islands, etc. then reinvest to approximative region where they actually operate, therefore, our research considers the approximative region investment as both direct and indirect. The investment which go through third region to approximative region would be considered as indirect approximative region investment and others are direct. Thus, the measurement of approximative region investment is the proportion from both direct and indirect approximative region investment to total investment which included inward investment. It’s calculated by dividing the sum of direct and indirect approximative region investment by total investment. Regions other than Taiwan, China, Hong Kong, Macao would be considered as non-approximative region for Taiwanese firms. The measurement of non-approximative region investment is the proportion of non-approximative region investment minus the indirect approximative region investment to total investment which included inward investment. It’s calculated by dividing non-approximative region investment minus indirect approximative region investment by total investment. (3)Moderator Internationalization Prior literatures measured it by the proportion of foreign sales to total sales (FSTS) (Geringer, Beamish & daCostsa, 1989; Geringer, Tallman & Olsen, 2000), proportion of foreign assets to total assets (FATA) (Daniels & Bracker, 1989; Ramaswany, 1995) and proportion of foreign employees to total employees (FETE) (Kim, Hwang & Burgers, 1989). According to UNCTAD, internationalization would measure by the sum from FATA, FSTS and FETE. But, in this research, we measure the degree of internationalization by the average of FSTS plus FATA because of data obtained. It’s calculated by dividing FSTS plus FATA by 2. Product Diversification In this study, we use Herfindahl index (Berry, 1975) to measure product diversification by summing the squared proportion of each product. The range of Herfindahl index are from 0 to 1 which mean lower product diversification while Herfindahl index close to 1, conversely, higher product diversification while Herfindahl index close to 0. (3) Control Variables 155
9. Table 3-3 Measure of Control Variables in This Study Variables Indicators Measurement Retain Earning Capital reserves/Total assets Control Variables Debt Ratio Total debts/Total assets Firm Size Natural logarithm of total assets R&D Intensity R&D expenses/Total sales Marketing Intensity Advertising expenses/Total sales 3.4.Descriptive Statistics & Correlation Matrix The statistical software employed is STATA v.12 which offers command to deal with generalized least square model and multilevel model and support graph drawing. The table of descriptive statistics of variables shows in the below Table 3-4A: Table 3-4A Descriptive Statistics Standard Variables Observations Mean Min Max Deviation ROA 6783 0.0432408 0.0817941 -0.7267451 0.597602 Approximative Region 6783 0.3839737 0.3425524 0 1 Investment Non-approximative 6783 0.1100567 0.2803171 0 1 Region Investment Internationalization 6783 0.4638367 0.2283915 0.0000256 0.8885477 Product Diversification 6783 0.501118 0.2294775 0.001024 1 Retained Earnings 6783 0.0844345 0.0916586 0 0.7752016 Debt Ratio 6783 0.4148845 0.1698312 0.0127107 0.9781516 Firm Size 6783 15.7313 1.645658 7.007653 21.62453 R&D Intensity 6783 0.0315692 0.0617839 0 0.95899 Marketing Intensity 6783 0.0543676 0.0545025 0 0.5120479 Table 3-4B Correlation Matrix 156
10. Table 3-4B above shows the Pearson correlation coefficient of each variables in this study. As the result, the coefficients of correlation are all less than 0.6 demonstrating there is no collinearity bias in our research. 4. Results The result of generalized least square model would show below table 4. Table 4 Result of GLS model 4.1. Relationship between Outward Investment and Performance From the result of Model 2, the hypothesis 1A (the relationship between approximative region investment and performance is significant inverted U-shape) is supported but the result also shows the insignificance at moderated model (Model 4, 5& 6). Thus, hypothesis would be confirmed partially supported. Figure 4-1A shows the relationship between them. The performance would increase with the level of approximative region investment, and start to decrease after 40% of investment level. The result of Model 2 examines that the additive relationship between non- approximative region investment and performance. Figure 4-1B shows the U-shaped, 157
11. curvy relationship between them. The performance would decrease at lower level of non- approximative region investment until 15%, after that performance would increase with the level of non-approximative region investment. Thus, the relationship between non- approximative region investment and performance is J-shaped not U-shaped. Furthermore, the result shows insignificance at moderated model (Model 4& 6). Therefore, the hypothesis 1B (the relationship between non-approximative region investment and performance is significant U-shape) is considered partially supported. .035 .04 .03 .035 ROA ROA .03 .025 .02 .025 0 .2 .4 .6 .8 1 0 .2 .4 .6 .8 1 Approximative Region Investment Non-approximative Region Investment Figure 4-1A Relationship between Figure 4-1B Relationship between non- approximative region investment and approximative region investment and ROA ROA 4.2. Effect of Internationalization to Outward Investment and Performance From the result of Model 4 and Model 6, internationalization does not moderate the relationship between approximative region investment and performance, conversely, it moderate the relationship between non-approximative region investment and performance. Hence, hypothesis 2 (the degree of internationalization would moderate the relationship between outward investment and performance) is only supported while non-approximative region investment occurs. Figure 4-2 shows the moderating effect of internationalization to the relationship between non-approximative region investment and performance. As the figure shown below, Better performance can be achieved in higher degree of internationalization than those in low degree and it would let the relationship between them be changed from J- shape to U-shape. The performance, for firms with higher internationalization would be improved after passing the 30% threshold of investment level, on the other hand, for firms with lower internationalization, the performance would be increased after 45% investment threshold level. 4.3. Effect of Product Diversification to Outward Investment and Performance The result of Model 5 and Model 6 suggest that there is insignificant effect of product diversification to non-approximative region investment and performance but significant effect to approximative region investment and performance. Thus, the hypothesis 3(the degree of product diversification would moderate the relationship 158
12. between outward investment and performance) is only supported while approximative region investment happens. Figure 4-3 shows the effect of product diversification to the relationship between approximative region investment and performance. It indicates that better performance can be achieved for firms with low product diversification than those in higher degree and product diversification would change the relationship of approximative region investment and ROA from inverted U-shape to U-shape. The performance of the firms with low product diversification would improve until the threshold level of 55%, afterward, the performance would increase. On the other hand, the performance of the firms with high product diversification would decrease until the threshold level of 55%, then increase. .07 .03 .065 .025 .06 ROA ROA .055 .02 .05 .045 .015 0 .2 .4 .6 .8 1 0 .2 .4 .6 .8 1 Non-approximative Region Investment Approximative Region Investment Low Intl High Intl Low PD High PD Figure 4-2 Effect of internationalization Figure 4-3 Effect of product to the relationship between non- diversification to the relationship approximative region investment and between approximative region ROA investment and ROA 4.4. Configuration of Outward Investment, Moderators and Performance 1. Configuration of Non-approximative Region Investment, Internationalization and Performance Figure 4-4A and Figure 4-4B below show the configuration of non-approximative region investment, internationalization and performance. It indicates that the firms with high degree of internationalization and low level of non-approximative region investment or high degree of internationalization and high level of non-approximative region investment would own the best performance (green contour line & red region), but the firms with low degree of internationalization and medium low level of non-approximative region investment would own the worst performance (purple contour line & navy blue region). 159
13. Figure 4-4A Configuration of Non- Figure 4-4B Configuration of Non- approximative Region Investment, approximative Region Investment, Internationalization and ROA in 3D Graph Internationalization and ROA in Surface Graph Figure 4-4D Configuration of Approximative Figure 4-4C Configuration of Approximative Region Region Investment, Product Diversification Investment, Product Diversification and ROA in 3D Graph and ROA in Surface Graph 2. Configuration of Approximative Region Investment, Product Diversification and Performance Figure 4-4C and Figure 4-4D below show the configuration of approximative region investment, product diversification and performance. It demonstrates that the firms with low level of product diversification and medium level of approximative region investment would own the best performance (green contour line & orange region), but the firms with high level of product diversification and medium level of approximative region investment would own the worst performance (black contour line & navy blue region). 4.5. Relationship of Outward Investment and Performance by MLM 160
14. Table 4-5 Result of MLM model This study applied MLM model for the purpose of examining the relationship of outward investment and performance by industry. The result of multilevel model demonstrate above at the table 4-5. By MLM model, its advantage is to display the effect of outward investment to performance by each industry. Figure 4-5A shows the relationship between approximative region investment and performance by each industries. The figure indicates that the % influence of ROA while increase a % of approximative region investment. According to the figure, there are mainly negative relationships between approximative region investment and ROA. The most serious industrial effect to ROA is optoelectronic industry, while 1% increase in approximative region investment would lead to 0.06% decrease in ROA. Conversely, the least effect is semiconductor industry. 161
15. Figure 4-5B which demonstrates the relationship between non-approximative region investment and performance by each industry indicates that optoelectronic industry is only one industry with negative relationship between non-approximative region investment and ROA. While 1% of non-approximative region investment increases, it would lead to 0.01% decrease in ROA from optoelectronic industry. However, the relationships between non-approximative region investment and ROA are positive by other industries, with the most serious effect from semiconductor industry and the least effect from computer and peripheral equipment industry. Cement Cement Food Food Plastic Plastic Textile Textile Electric Machinery Electric Machinery Electrical & Cable Electrical & Cable Paper & Pulp Paper & Pulp Iron & Steel Iron & Steel Rubber Rubber Automobile Automobile Other Manufacturer Other Manufacturer Chemical Chemical Biotechnology & Medical Care Biotechnology & Medical Care Semiconductor Semiconductor Computer & Peripheral Equipment Computer & Peripheral Equipment Optoelectronic Optoelectronic Electronic Parts/Components Electronic Parts/Components Other Electronic Other Electronic -.01 0 .01 .02 -.06 -.04 -.02 0 % change of ROA, with % increase on NARI % change of ROA, with % increase on ARI Figure 4-5A Effect of industries on the Figure 4-5B Effect of industries on the relationship between approximative region relationship between non-approximative investment and ROA region investment and ROA 5. Conclusion 5.1. Conclusion and Discussion To sum up the result of this study, the relationship between approximative region investment and performance is inverted U-shaped (Figure 4-1A). While the firms invest to the approximative region which can be relatively easier to profit from lower liability of foreignness, as lower risks of similar context, combined with new markets and new resources cause performance be increased. However, after the investment achieve a specific threshold level (40%), firms would encounter not only a phenomenon of the saturation of favorable markets, but the competition from competitors or emerging local firms, and the instability of host countries which lead to decreased performance (Gomes & Ramaswamy, 1999). Hence, the dependence of approximative region investment shouldn’t be too high. It implies why the proportion of investment to China is relatively high from Taiwanese firms. The business context and operation system of Chinese region are similar to Taiwan, causing lower extra costs and risks but benefits to invest for Taiwanese firms. But, the performance would be decreased while the investment achieve a specific level. Therefore, this study suggest that it should invest to the region with familiar context and lower risks without over concentration. 162
16. Table 5-1 Result of Hypothesis Hypothesis Content Result Hypothesis The relationship between Partially Supported 1A approximative region investment and performance is significant inverted U- shape. Hypothesis The relationship between non- Partially Supported 1B approximative region investment and (Result is J-shape) performance is significant U-shape. Hypothesis The degree of internationalization While approximative region 2 would moderate the relationship investment: Not Supported between outward investment and While non-approximative region performance. investment: Supported Hypothesis The degree of product diversification While approximative region 3 would moderate the relationship investment: Supported between outward investment and While non-approximative region performance. investment: Not Supported The relationship between non-approximative region investment and performance is J- shaped (Figure 4-1B). The performance would be decreased because of lack of knowledge for operation by different context while the firms invest to the region with high liability of foreignness. There would be the extra “tuition” for firms causing costs larger than benefits, until a specific level of investment. After that investment level, firms already gradually accumulate the experience and familiarize that different context causing performance beginning to increase. Thus, the relationship between them should be U-shaped. However, the probable reason of difference between result and hypothesis are the support of home country’s government and the welcome of host countries. Recently, many governments encourage firms to conduct outward investment, additionally, many emerging countries welcome the FDI. They give special treatment and support to the firms causing the costs of liability of foreign decreased. As the result, after the investment level achieve 15%, the performance would start to increase (J-shape). This empirical result give the new choice for Taiwanese firms which consider the risks and costs are huge while conduct non- approximative region investment. In sum, the liability of foreign costs is not huge as imagination, firms should premeditate other region for investment except for Chinese region. From the result, internationalization wouldn’t affect to the relationship between approximative region investment and performance. The reason of it might be that threshold of approximative region investment for firms with high internationalization or low internationalization is same. For firms, regardless of the degree of internationalization, the contextual risks and costs of approximative region are similar resulting in the insignificant effect of internationalization to approximative region investment and performance. Moreover, internationalization significantly affect to the relationship between non- approximative region investment and performance, better performance for the firms with 163
17. higher degree of internationalization, additionally, the relationship between them would be changed from J-shape to U-shape by the lower degree of internationalization (Figure 4-2). While the firms in lower degree of internationalization conduct non-approximative region investment, more costs from liability of foreignness for firms lead the performance to be decreased until the investment level achieve 45%. It’s necessary for firms in low internationalization to increase more non-approximative region investment before the performance start to increase. Conversely, the firms in higher internationalization would own more international experience and resources, which could decrease the liability of foreignness while the firms conduct non-approximative region investment causing the relationship between non-approximative region investment and performance would maintain J-shape and better performance. The result shows insignificant effect of product diversification to the relationship between non-approximative region investment and performance. Outward investment is an instrument for firms to internationalize, however, the firms conduct geography diversification wouldn’t conduct product diversification because of limited resources (Chatterjee, Wernerfelt, 1991). In other words, firms couldn’t conduct international diversification and product diversification simultaneously because firms only could place finite resources to one strategy. According to prior research, the relationship between international and product diversification also was insignificant (Geringer, Tallman, & Olsen,2000；Tallman & Li, 1996). Conversely, the risks and costs of conducting approximative region investment is relatively less, causing it is not necessary to employ more resources to conduct it. Thus, it is probable for firms to conduct them at the same time. Different product diversification are significantly affect to the relationship between approximative region investment and performance. Better performance can be achieved for the firms with low product diversification and high degree of product diversification would change the relationship between them from inverted U-shape to U-shape (Figure4-3). Costs for developing new markets and new business are larger than benefits for the firms with high degree of product diversification while conduct approximative region investment. After 60% of investment level, performance would be increased by developed new markets and products. Conversely, the relationship between approximative region investment and performance would be maintain to inverted U-shape by the effect of low product diversification and be better performance. Figure 4-4A & B display that the best performance is happened while the firms with high degree of internationalization and low or high level of non-approximative region investment. It indicates that it’s better for firms to operate internationally and conduct concentrative investment at non-approximative or approximative region which equals low level of non-approximative region. If firms with high internationalization conduct approximative and non-approximative region investment simultaneously, due to limited resources, it might increase the transfer and managerial costs of resources which cause decreased performance. Therefore, it propose that the firms with high degree of internationalization should conduct outward investment at concentrative region. On the other hand, there would be the worst performance while the firm at low degree of 164
18. internationalization and medium low level of non-approximative region investment, because low degree of internationalization would cause limited markets and resources and it’s necessary for firms to pay the tuition while initially conduct non-approximative region investment. Thus, the worst performance happen. In conclusion, better performance would occur by high internationalization, but awareness of proportion between approximative and non-approximative region investment is required. Figure 4-4C & D show that the best performance would occur, while the firms with low product diversification and medium approximative region investment. Firms with low product diversification conduct approximative region investment, they would spend less extra cost for product development, but obtain new markets and resources from familiar place causing the best performance happened at medium level of approximative region investment. On the other hand, the worst performance would occur while firms conduct medium level of approximative region investment with high product diversification, because there might be not only extra product developing costs but also market developing costs. For firms with high product diversification, developing market for diversified- product which cause extra cost is required. Therefore, it suggests that it’s more suitable for firms with low product diversification to conduct approximative region investment without over-concentrating. As the result from MLM, the relationships between approximative region investment and performance are negative by each industry (Figure 4-5A). Most serious negative influence is from optoelectronic industry, the influence of ROA would be negative 0.06% with 1% increasing of approximative region investment. Conversely, the influence of ROA would be negative 0.04% with 1% increasing of approximative region investment by semiconductor industry. In sum, while semiconductor industry conduct approximative region investment, it would own the best performance but optoelectronic would own the worst. Furthermore, positive relationship between non-approximative region investment and performance would be happened by each industry except for optoelectronic industry (Figure 4-5B).The most serious positive influence is from semiconductor industry, the influence of ROA would be positive 0.02% with 1% increasing of non-approximative region investment, on the other hand, it would be negative 0.01% to ROA with increasing of non-approximative region investment by optoelectronic industry. Thus, semiconductor industry conduct non-approximative region investment would occur the best performance and the worst would be from optoelectronic industry. In conclusion, the suggestion of this study is that each industries could obtain positive influence by non-approximative region investment and approximative region investment is still necessary according to Defensive Investment (Kojima, 1978 & Ozawa, 1979) which considered some of investment couldn’t obtain profit, but it’s still necessary to invest because of competitors. In addition, semiconductor industry is the most suitable to conduct outward investment. But for optoelectronic industry, no matter approximative or non-approximative region, there are just negative influence. Therefore, it’s not necessary for optoelectronic industry to conduct outward investment. 165
19. 5.2. Limitation Precise region distinction of in accordance to liability of foreignness is not possible in this study which just distinguish low liability of foreignness to approximative region and others are non-approximative regions causing it couldn’t examine relationship while the region with medium level of liability of foreignness. It’s necessary for future research to consider it. REFERENCE 1. Ajais. Gaur; Vikas Kumar; Ravis Sarathy (2011).Advances in International Management: The Past, Present and Future of International Business and Management, 24 2. Aulakh, P. S., Kotabe, M., & Teegen, H. (2000). EXPORT STRATEGIES AND PERFORMANCE OF FIRMS FROM EMERGING ECONOMIES: EVIDENCE FROM BRAZIL, CHILE, AND MEXICO. Academy of Management journal, 43(3), 342-361. 3. Barbosa, N., & Louri, H. (2005). Corporate performance: Does ownership matter? A comparison of foreign- and domestic-owned firms in Greece and Portugal. Review of Industrial Organization, 27, 73-102. 4. Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120. 5. Chhibber, P. K., & Majumdar, K. K. (1999). Foreign ownership and profitability: property rights, control and the performance of firms in Indian Industry. Journal of Law and Economics, 42, 209-238. 6. Daniels, J. D., & Bracker, J. (1989). Profit performance: Do foreign operations make a difference? Management International Review, 29, 46-56. 7. Dunning, J. H., 1993, Multinational Enterprises and the Global Economy, Addison 2 Wesley Publishing Co. 8. Geringer, J. M., Tallman, S. & Olsen, D. M. (2000). Product and international diversification among Japanese multinational firms. Strategic Management Journal, 21(1), 51-80. 9. Geringer, J. M., Beamish, P. W., & daCosta, R. C. (1989). Diversification strategy and internationalization: Implication for MNE performance. Strategy Management Journal, 10, 109-110. 10. Grant, R. M., Jammine, A.P., & Thomas, H. (1988). Diversity, diversification, and profitability among British manufacturing companies. Academy of Management Journal, 31, 771- 801. 11. Hitt , M. A. , Uhlenbruck , K. and Shimizu , K. , 2006 ,“The Importance of Resources in the Internationalization of Professional Service Firms : the Good , the Bad , and the Ugly”, Academy of Management Journal , 49 , 1137 —1157. 166
20. 12. Hymer, S. H. (1976). The international operations of national firms: A study of direct investment. Cambridge, MA: MIT Press 13. Kim, W., Hwang, P., & Burgers, W. (1989). Global diversification strategy and corporate profit performance. Strategic Management Journal, 10, 45-57. 14. Kotabe, M. (1990). The relationship between offshore sourcing and innovativeness of U.S.: An empirical investigation. Journal of International Business Studies, Fourth Quarter, 623-638. 15. Lu, J., & Beamish, P. (2001). The internationalization and performance of SMEs. Strategic Management Journal, 22, 565-586. 16. Lubatkin, M. H., Chatterjee, S. (1994). Extending modern portfolio theory into the domain of corporate diversification: Does it apply? Academy of Management Journal, 37, 109-136. 17. Pankaj Ghemawat, (2001). Distance still matters. Harvard Business Review 79(8). 18. Ramanujam, V., & Varadarajan P. (1989). Research on corporate diversification: A synthesis. Strategic Management Journal, 10(6), 523-551. 19. Ramaswami, S. N., Srivastava, R. K., & Bhargava, M. (2009). Market-based capabilities and financial performance of firms: Insights into marketing's contribution to firm value. Journal of the Academy of Marketing Science, 37(2), 97-116. 20. Ruigrok, W., & Wagner, H. (2003). Internationalization and performance: An organizational learning perspective. Management International Review, 43, 63-84 21. Shleifer, A., & Vishny, R. (1990a). Managerial entrenchment, the case of manager specific investments. Journal of Financial Economics, 25, 123-139. 22. Siddharthan N. S., & Lall, S. (1982). The recent growth of the largest U.S. multinationals. Oxford Bulletin of Economics and Statistics, 44 (1), 1-13. 23. Tallman, S., & Li, J. (1996). Effects of international diversity and product diversity on the performance of multinational firms. Academy of Management Journal, 39(1), 179-196. 24. Teece, D. J. (1982). Toward an economic theory of Muiltiproduct firm. Journal of Economic Behavior and Organization, 3(1), 39-63. 25. Vernon, R. (1971). Sovereignty at bay: the multinational spread of US enterprises. Basic Books, New York. 26. Williamson, Oliver E. (1975). Markets and hierarchies: Analysis and antitrust implications. (Free Press, New York). 27. Zaheer, S. (1995). Overcoming the liability of foreignness. Academy of Management Journal, 38 (2), 341-364. 167