How do domestic owned firms and foreign owned firms differ in terms of performance in different industries of China?
Master Thesis
Student: Li Cao Student ID: s1400088 Email: singinger@hotmail.com Supervisor: Dr. Jianhong. Zhang Co-assessor: Prof. dr. H. van Ees MSc. International Economics & Business
Faculty of Economics
RijksUniversiteit Groningen, The Netherlands
July 26th, 2007
ACKNOWLEDGEMENT
How time passes by, my four year study in the Netherlands comes to an end. I really appreciate all the teachers and tutors who have coached my IE&B program study these years. What RuG provides me is not only the knowledge but also the enthusiasm of obtaining knowledge. I am so contented with what I have got these years in the Netherlands, by which my knowledge is enriched and my horizon is broadened. It is definitely one of the best memories in my whole life.
Concerning this research, I thank to all those who gave me the possibility to complete this thesis. I do appreciate the assisting efforts from my supervisors: Ms. Zhang and Prof. Van Ees. Without your encouragements and inspiring suggestions, I can not imagine I could finish this thesis in time. In addition, I would like to express my gratitude to my parents whose patient love has strongly supported me these years.
ABSTRACT
This paper explains to what extent do domestic and foreign owned firm differ from each other in performance of different industries in China. Industries are classified according to its nature resource abundance and pollution intensity, performance is investigated in terms of profit. Ordinary Least Squares regression models with ownership as dummy variable and types of industry as nominal variables are tested. The results show that domestic ownership has a negative influence on firm performance, and foreign owned firms perform better than domestic firms under all types of industries.
KEY WORDS
China, Domestic owned firms, Foreign owned firms, Performance, Industry, Nature resource based, Pollution intensive
JEL Classification: F14, F23
TABLE OF CONTENTS
1. INTRODUCTION... 5
2. LITERATURE REVIEW ... 8
3. DATA AND METHODOLOGY ... 24
DATA COLLECTION... 27
METHODOLOGY... 29
Dependent variable ... 29
Independent variables – O, TOI... 29
Control variables – ABCF, GIOV, CE, TSEC ... 30
Description of the analysis... 30
4. RESULTS AND DISCUSSION ... 37
5. CONCLUSION ... 42
REFERENCES... 43
APPENDIX ... 47
1. INTRODUCTION
Till the end of year 2006, it has been eighteen years since the provoke of privatization in China and five years since China’s entry to the World Trade Organization (WTO).
Privatization process by the late 1970s not only turned the “State Owned Enterprise” to become “State Run Enterprise”, but also attracted huge foreign investments to mainland China. Seeing the yearly increasing GDP rate and the positive growth anticipations, it has no doubt that China has accumulated enormous capital by exports and FDI these years.
Capital accumulation could be one reason to explain the increasing of domestic wages and appreciation of Chinese Yuan against US dollar in 2006 (Adams, et. al 2006). It seems that labor cost is no longer cheap compared to other neighborhood countries such as Vietnam. Questions therefore arise by wondering if China is still the “manufacturing factory” of the world in the next a few decades. Then, if China is no longer attractive to foreigners for its cheap labor force, what will be the consequences? Does this mean that China will switch to become a rent seeker towards other developing regions, for instance Africa?
From economic theories we shall say that one condition for China to become a rent seeker is that local firms already perform better than foreigners in the domestic market, which gives evidence that the Chinese firms have their own advantages in the international rivalry battle. However, most studies of multinationals in the host economy have shown that foreign owned firms achieve a superior performance compared with domestic owned firms. This is also referred to as a so called “new trade theory” by economists such as Markusen and Venables (1997). By investigating the effect of ownership on post-privatization performance for some Central East European countries, it is also found that foreign ownership improves economic performance, but domestic ownership does not (Kocenda and Svejnar, 2003). Likewise, previous studies regarding FDI in China addressed the industrial development and concluded that multinationals have the highest sales profit rate, (Wang, 1997) and have impacts on the Chinese economic structure different from that of domestic firms and other foreign invested firm (mainly Hong Kong, Taiwan and Macao firms) (Zhang, 1998). It seems that foreign firms
have done better than domestic owned firms in past years (OECD, 2000). However, if we look at an industry level, the exact performance differences between foreign firms and national firms are still not clear. OECD (2000) found that foreign owned enterprises overtook domestic owned enterprise as main producers of electronic goods, cultural goods, sports goods and leather products. While in thirteen other manufacture sectors, domestic owned firms (mainly collective and private enterprises) were responsible for more than half of industrial production. In these sectors foreign owned enterprises were less in competition with domestic owned firms in terms of profit.
Nevertheless, former studies on domestic and foreign competition in the Chinese economy have several limitations: firstly, most earlier researches only looked at the international market for exporting trading volume. Seldom have done for the host market competition on a profit level. Secondly, not only the sound international researches but also the regional Chinese studies are quite a few years ago. Studies for China after entering WTO in 2001 are rather limited. Thirdly, past studies didn’t indicate the extent to which performance of foreigners and locals differ from each other, for instance, there lacks of explanations at an industry level. Therefore, hopefully this paper can cover those gaps by answering one general research question of:
To what extent do domestic owned firms and foreign owned firm differ from each other in terms of performance in difference industries of China?
This study will focus on the recent Chinese industry sectors, performance of domestic and foreign owned firms will be checked in terms of profit, along with a few rational hypotheses derived form different academic voices. It can give us a picture of which types of industries the domestic owned firms have competitive advantages in, while which types of industries the foreign owned firms do better. It can also show that after eighteen years of privatization and five years since join into WTO, whether domestic firms have gained from participation in international activities or not. Last but not least, this work is of great interest to potential investors, and can serve as a good predictor for the Chinese industries in the future.
As part one gives the introduction, the structure for the rest of the paper is: Part two contains the literatures related to performance comparisons of domestic and foreign
enterprise, specific research questions, hypotheses, and theories for the control variables of the testing regression. Part three explains the data and methodology. The methodology is based on an ordinal least square regression model that sets ownership and types of industry as dummy and nominal variables, plus a few control variables. Part four discusses the findings, results and limitations. Conclusion will be made in the last part.
2. LITERATURE REVIEW
Prior to the late 1990s, multinational corporations (MNCs) entered to the Chinese market and acted merely as “foreign investors”. “Foreign investors” operated in only a single or a few projects, often by applying a low risk entry mode to test the market. They located in major coastal cities, and targeted 1st-tier regional consumers for their core products.
Since China’s entry to the WTO from 2000, more opportunities are opened to foreign firms. For these who have already been in China for a long time, their first mover advantages allow them to grow at a sustainable rate. Many MNCs have evolutionarily turned to become “strategic insiders”. “Strategic insiders” conduct multiple businesses in multiple locations, operate on an extremely large scale, and with various entry modes including high risk ones such as acquisition (Luo, 2006). They coordinate operations through their Chinese centers or subsidiaries, while maintain both organizational control and strategic flexibility. However, MNCs face handicaps in competition with local producers, including poor supporting infrastructures; fragmented distribution channels;
inadequate flexibility; higher costs and protected trade barriers across provinces (Luo, 2006). Therefore, what drives the financial performance of MNCs’ oversea operations has long been a topic of great interest to academia (Douglas and Craig, 1983; Guisinger, 1989; Osland and Cavusgil, 1996). And this triggers the first motivation of this research, which is to explore the performance of domestic owned firms in comparison with foreign owned firms. Besides, new generation of MNCs have markedly different opportunities and threats across different industries in China. For instance, companies selling products such as elevators, subway systems, automobiles and mobile phones found China their most critical market for growth. While companies selling products such as consumer electronics, processed foods and pharmaceuticals find that they have already missed the best entry timing (Luo, 2006). So the second motivation of this paper is to discover the performance of domestic owned firms and foreign owned firms in different industries of China.
First of all, it is interesting to know the reasons why firms are eager to invest in another country. Answers regarding the entrance of foreign firms are mostly explained by
Dunning’s OLI theory, which addresses the different ownership, location and internalizing (OLI) advantages to multinational companies.As Dunning (2001) puts it,
“OLI framework basically states that the extent and the pattern of multinational operations are determined by firm specific ownership structure, geographic location and internalizing priorities.” Although it is still questionable of how far MNCs’ OLI advantage are applicable in transitional markets such as China, researches show that OLI advantages have been assisting China’s economic restructuring towards higher allocative and technical efficiency (Zhang, 1998). Among the three OLI variables, ownership advantage is the strongest motivation to attract foreign investments. Moreover, ownership-specific advantages are the strengths that MNCs can exploit to enhance their asset power. As Dunning (2001) says that a key goal of asset seeking investment is to protect the investing firm’s core competences, including the access to new resources and capabilities, and the ability of the investing firm to manage these ownership advantages.
It has found that the imports of equipment and new technology are the main determinants of ownership-specific advantages for foreign firms in China. MNCs establish sub- companies in China to utilize their technology advantages so as to compete with local enterprises effectively (Mao and Luo, 2001). Although the theory of asset-exploiting MNC activity is still in its infancy, it seems likely to challenge researchers in the next decade (Dunning, 2001).
As lessons can be learned from China’s economic reform over 20 years, one might ask what the outcomes of the competition between domestic and foreign owned firms are.
And who are the gainers while who are the losers? Theoretically, there are no uniform answers. Some say that given a country and its industry contexts, firms with a higher proportion of foreign ownership will on average perform better than their domestic counterparts (Caves, 1982). Multinational firms are deemed to be significantly different from national firms, as foreign investors are assumed to bring in bundles of things that are not available locally. The entrance of a multinational firm in the product market might simply crowed out national firms and compete away their market shares (Navaretti and Venables, 2003). Furthermore, as foreign entrants can possess OLI advantages over domestic competitors, which will also ensure them to record a superior performance
(Szymanski, et. al, 2006). However, the benefits of domestic owned firms can also take place through several different channels. One is that investment makes the market more competitive, eroding monopoly power of local firms. Another is that investment raises the productivity of local firms through spillover effect. This happens if increased competitive pressure induces firms to reduce internal inefficiencies, or if there are direct knowledge spillovers or learning effects. For instance, the presence of foreign firms might enable local firms to learn new technologies, management methods and market opportunities (Navaretti and Venables, 2003).
Noticing that there are opposite effects, research on MNCs is often named as “conditional approach” which states that foreign ownership improves firm performance, but no evidence have been found for what the exact differences in performance between multinational firms and national firms are. The effect on performance becomes much weaker and in some cases is not even significant.1 Besides, some criticisms say that this model is implicit as most of the studies are based on UK and US affiants in developed countries. For countries in transitional economies, researches on firms in India indicate that the positive effect of foreign ownership on firm performance is substantially attributable to foreign corporations that have, on average, larger shareholding, higher commitment and longer-term involvement (Douma et. al, 2003). Studies on Czech Republic’s medium and large firms show that concentrated foreign ownership improves economic performance, but domestic private ownership does not2 (Kocenda and Svejnar, 2003). Back to the questions of China, generally speaking, the effect of international ownership is unpredictable and domestic institutional ownership does not appear to improve performance. Pure domestic ownership sometimes even has a negative effect on firm value (Wei and Varela, 2002), while foreign owned firms, especially those who have an equity joint venture with local firms, have a highest profit level among all types of firms (Pan and Chi, 1999).
1 Researches on the effect of performance will be explained more in the methodology part of this paper.
2 Kocenda and Svejnar (2003) further say that “Foreign firms engage in strategic restructuring by increasing profit and sales, while domestic firms reduce sales and labor cost without increasing profit.”
Although hardly any researches have found the precise differences between domestic and foreign owned firms with respect to their performance. Statistical tests done by OECD3 demonstrate that foreign capital is a main engine to trigger endogenous growth in the Chinese manufacturing industries. There is a positive relationship for the share of foreign capital and the annual growth rate of industrial production between 1994 and 1997. This positive relationship indicates that sectors with a better endowment of foreign capital grew most rapidly during the 1994-97 period (OECD, 2000), although there have been some exceptions.4 Likewise, the overview of China’s absorption of FDI report (MOFCOM, 2005) says that the main area of FDI inflows are towards the manufacturing sectors, in which the sum of foreign-invested enterprises account for 72.03% of the total accumulated volume. Hong Kong, Macau and Taiwan province accounted for 33.43%
FDI inflows in China while other foreign countries accounts for the rest. Seeing these empirical findings, it is conclusive to say that the foreign capitals do play a powerful role in the Chinese manufacturing industries; the significances of the OLI advantages are vary across industries; and the differences among industries are important factors to determine profitability. Based on these ideas, this research will focus on comparing domestic and foreign owned firms at an aggregate industrial level. To see the actual links between ownership and performance, I firstly expand the general research question by trying to clarify,
1 a. How can domestic ownership influence firm performance at an industrial level in China?
1 b. How can foreign ownership influence firm performance at an industrial level in China?
In order to estimate the performance of both ownerships, it is necessary to firstly define different types of industries. The traditional segmentation method divides industries into the primary sector of industry, including agriculture, mining and raw material extraction;
the secondary sector of industry, which normally stands for manufacturing; and the tertiary sector of industry, which is service production. Another criteria to define industry
3 OECD stands for the Organization for Economic Co-operation and Development
4 For instance, Wang (1997) found that MNCs have the highest sales profit rate (6.37%), while sales profit
is simply by light industry and heavy industry. The third way to classify industry is by looking at the final products stage.5 China Statistical Year Book (2006) adopts the third method and industries are allocated according to their respective final outputs. Inspire by the theory of Schumacher (1973) which says that economic performance needs to be questioned on the availability of basic resources and on the capacity of the environment to cope with the interference implied. This paper will develop this idea of resources(inputs)-environment(outputs) relation and take all manufacturing industries from China Statistical Year Book (2006).
From an input perspective: nature resources based sectors
The motivation of this research to study the nature resources based sector in the Chinese industry is because natural resources make important contributions to long-term economic productivity, especially for developing countries, which are usually dependent on their nature resources base for employment and exports (Repetto, 1989). The general accepted terms of natural resources are referred to as: naturally occurring substances that are considered valuable in their relatively unmodified or natural form (Schumacher, 1973). According to China Statistical Yearbook (2006), three broad nature resources are land resources, forest resources and water resources. Furthermore, at an industrial level, agricultural, forestry mining, and petroleum extraction are categorized as natural-resource industries. 6 The rest of them are considered as non-nature resource industries. Public policy plays a disproportionately large part in determining the economic success of nature resource based industry7 (Schloss, 2000). E.g. statistical results from OECD (2000) found
5 According to ISIC, rev.4, this includes chemical industry, petroleum industry, meatpacking industry, hospitality industry, food industry, fish industry, software industry, paper industry, entertainment industry, semiconductor industry, cultural industry, and poverty industry. ISIC stands for International Standard Industrial Classification of ALL economic activities, the most complete and systematic industrial classification made by United Nations Statistics Division.
6 Although some categorizing facilities, such as the nature resources department of Canada (see, http://www.nrcan-rncan.gc.ca/com/) excludes agricultural from the sector of nature resources, the China Statistical Yearbook (2006) indicates that figures on area of cultivated land of the land sector were taken from the 1996's agricultural census.
7 It is furthermore recognized that within the natural resource industries, there is always a serious reputation problem of corruption. The depth of this problem is accurately captured by the way in which unsustainable overexploitation of forests, fisheries or farmlands is commonly referred to as mining the forests, mining the seas, mining soils, and more broadly mining countries. For instance, more than most industries, mining relies on a high level of public consent in order to continue its activities, since states tend to exercise a significant degree of control over access to and exploitation of mineral resources.
that Stated Owned Enterprise (SOEs) represented more than 55 % of output in mainly 6 sectors (tobacco, timber, petroleum and gas extraction, petroleum processing, coal mining, ferrous metallurgy) of year 1997 in China.
From an output perspective: pollution intensive sectors
The motivation of this research to study the pollution intensity sector in the Chinese industry is that environmental regulations, which are mainly designed exogenously by both domestic and international policy makers, play an important role in affecting economic performance. Ambitious environmental policies could harm a nation’s comparative advantage, reflecting a long-term movement of manufacturing capacity from the domestic country to other countries, in particularly, to the “pollution havens” (Jaffe et.
al, 1995). The “pollution haven hypothesis” indicates that, countries set lax environmental controls as their comparative advantage to attract FDI. Open countries who have a particularly high demand for environmental quality (rich countries) specialize in products that can be produced cleanly, letting poor open countries produce and sell the products that require pollution (Frankel and Rose, 2005). Restricted by the harsh pollution rules of the home country, investors, in most case, entrepreneurs from developed countries, will seek business opportunities in developing countries where pollution regulation is loose. There are various measurements of pollution intensity. The first one is by the aggregate toxic releases per unit of output: metals, cement, pulp, paper, and chemicals are identified as the dirtiest industries. The second one is by the emission intensity, which is the actual emissions per unit of output. The five most pollution intensive industries are: iron and steel, non-ferrous metals, industrial chemicals, pulp and paper, and non-metallic minerals.8 Besides these two measurements, oil, natural gas, and coalbed methane extraction are considered to be potential point sources of water pollution (Gallagher and Ackerman, 2000).
This research will study the joint effects of nature resources and pollution intensive features on each industry. The motivations are: firstly, economists have merely paid any attentions on the effects of environmental regulation to the competitiveness in the natural resources sector. There is an absence of environmental regulation on the manufacturing
8 The five cleanest industries, by the same standard, are textiles, non-electrical machinery, electrical
of natural resource industries such as forestry, agriculture, mining, and commercial fishing.9 Since natural capital is immobile even in the long run, environmental regulations pertaining to natural resource industries do affect where the industries are located (Jaffe et. al, 1995). Second, FDI decisions in the “dirty industries” is also tightly connected to the location of nature resources, yet these raw material resources are located in both rich and poor countries. It is impossible to measure any impacts that lax environmental regulations without the supplementary of nature resources (Clapp, 2002). Thirdly, natural resource endowments may partly explain the pattern of pollution-intensive outputs. Based on the theory of comparative advantages, it is known that countries produce a high proportion of pollution-intensive goods may do so because their natural resource base makes them efficient producers of particular pollution intensive products.
Seeing these un-neglectful relations, I draw an industry segmentation matrix based on nature resource/non-nature resource and pollution intensive/non pollution intensive as the guideline for the research part. All industrial divisions, as picked up from China Statistical Yearbook (2006), are further categorized under their respective features of nature resources endowment and pollution intensity.
9 Jaffe et. al, (1995) further indicate that “Similarly, regulations pertaining to pesticide use in agriculture, the reclamation of land mined for coal or non-fuel minerals, or the equipment that can be used by commercial fishing fleets can clearly affect the costs faced by (and hence the international competitiveness of) firms in these industries. ”
Table 1. Industry segmentation matrix
Nature Resource Based Industry Non-nature Resource Based Industry
Pollution Intensive Industry
Type I
• Mining and washing of coal
• Extraction of petroleum and natural gas
• Mining and processing of ferrous metal ores
• Mining and processing of non-ferrous metal ores
• Mining and processing of nonmetal ores
• Mining of other ores
• Manufacture of non-metallic mineral products
• Smelting and pressing of ferrous metals
• Smelting and pressing of non-ferrous metals
• Manufacture of metal products
• Processing of petroleum, coking, processing of nuclear fuel
• Production and supply of gas
• Manufacture of paper and paper products
Type II
• Manufacture of chemical fibers
• Manufacture of raw chemical materials and chemical products
• Manufacture of medicines
• Recycling and disposal of waste
• Manufacture of plastics
• Manufacture of rubber
Non-pollution Intensive Industry
Type III
• Processing of timber, manufacture of wood, bamboo, rattan, palm and straw products
• Manufacture of furniture
• Processing of food from agricultural products
• Manufacture of foods
• Manufacture of beverages
• Manufacture of tobacco
Type IV
• Manufacture of general purpose machinery
• Manufacture of special purpose machinery
• Manufacture of transport equipment
• Manufacture of electronically machinery and equipment
• Manufacture of communication equipment, computers and other electronic equipment
• Manufacture of measuring instruments and machinery for cultural activity and office work
• Manufacture of artwork and other manufacturing
• Manufacture of textile
• Manufacture of textile wearing apparel, footwear, and caps
• Manufacture of leather, fur, feather and related products
• Manufacture of articles for culture, education and sport activity
• Printing, reproduction of recording media
• Production and supply of electric power and heat power
• Production and supply of water Source: China Statistical Yearbook (2006)
As can be seen from the industrial segmentation matrix in table 1, there are four types of industries. For each type of industry, it is attempted to discover their respective relations with different ownerships. Accordingly, the second specific research questions are:
2 a. What are the relations of ownership on the performance of firms in nature resources based and pollution intensive industries (type I) in China?
2 b. What are the relations of ownership on the performance of firms in nature resources based and non-pollution intensive industries (type II) in China?
2 c. What are the relations of ownership on the performance of firms in non-nature resources based and pollution intensive industries(type III) in China?
2 d. What are the relations of ownership on the performance of firms in non-nature resources based and non-pollution intensive industries(type IV) in China?
Nature resources based and pollution intensive industries (type I)
An inspection at the nature resources based and pollution intensive industries in table 1 gives us the idea that they are exact what “dirtiest industries” mentioned in the early part.
On the one hand, from a comparative advantage theory which states that a country should specialize in products in which they have a comparative advantage (Ricardo, 1974), we know that the developing countries are more likely to gain in the production of pollution intensive goods than in clean ones. Firms which have the right to pollute freely may thus have their comparative advantages in that industry (Jaffe et. al, 1995). However, the rapid growth of foreign trade in China seem accelerate the development of environment protection to a certain extent. Whether or not pollution havens exist in developing countries such as China is still subject of much debate (Clapp, 2002). As strengthening environmental management can bring tremendous trade opportunities to many more products, especially environmentally friendly products. Besides, China’s membership in WTO is conducive to industry restructuring and pollution intensity products controlling (Ye, 2000). On the other hand, natural resource endowments are particularly important determinants of the volumes of production (Leamer, 1984). As in Russia capitalism, the oligarchs act as a form of government enterprise, control most of the nature resources and make business from them (Grieve and Rachinsky, 2005). Similarly, it is also reasonable
to predict that DOFs have exclusive rights for the production of nature resource based industries, which are determined by China’s ecological factors and politicians. In addition, when a firm pollutes, it is essentially using a natural resource (a clean environment). And when a firm is compelled to reduce its pollutant emissions, that firm has, in effect, seen its access to an important natural resource reduced. Seeing these conflict opinions, the first hypothesis is,
H 1: Domestic owned firms perform better than foreign owned firms in the nature resources based and pollution intensive industries in China.
Nature resources based and non-pollution intensive industries (type II)
Table 1 shows this category contains agricultural and forestry products. Governments in almost every country, whether it is economically developed or still developing, have the incentive to protect their local agricultural industries. They do this to ensure a solid tax base and maintain employment, local governments can erect various barriers of trade to protect local industries from foreign competition (Bai et al., 2003). This is especially the case for China, where agriculture is the most important economic sector, employing over 300 million farmers (nearly half of the total work force). The forestry reserve in China is also quite considerable, covering around fifth of China’s land area. Inspired by the comparative advantage theory again, China is assumed to specialize in agricultural and forestry products which it is well endowed. Consequently, DOFs have their competitive advantages in this agricultural and forestry industry. Possible reasons could be either the long tradition of production in this industry type or the local government protectionism similar as in type I industry. Empirical evidence also gives evidences that there are relatively lower tax burden to domestic producers than foreign producers in some agricultural products, such as beverage and tobacco (OECD, 2000). Summarizing these discussions, the second hypothesis is,
H 2: Domestic owned firms perform better than foreign owned firms in the nature resources based and non-pollution intensive industries in China.
Non-nature resources based and pollution intensive industries (type III)
Table 1 shows this sector is chemical related industries. A typical characteristic of chemical industry is its knowledge intensity. Zhang (1998) found that MNCs focus more
on knowledge intensive sectors in China. Although some argue that the possibilities of direct knowledge spillovers are obtainable though the learning of domestic enterprises.
As a result, local firms are able to increase their productivity level through this spillover effect. Still, knowledge intensive industries invested by foreigners will experience more advantages than local producers because of their more advanced technology capital.
Empirically, one report from Deutsche Bank regarding the chemical industry in China states that western multinationals and local Chinese companies are currently setting up extensive production capacities in China.10 Direct investment plays a key role in the chemical industry. But the increase of capacity is mainly being financed by direct investments as part of cooperative ventures with foreign business partners. I then provide the third hypothesis as,
H 3: Foreign owned firms perform better than domestic owned firms in the non-nature resource based and pollution intensive industries in China.
Non-nature resources based and non-pollution intensive industries (type IV)
This type contains various manufacturing sectors. It is possible to groups them into four major sectors: machinery and equipment, consumer electronics, textile products, power and water, and the others. Firstly, new technologies play an important role in determining the production of machinery and equipment. As mentioned in the literature review part, technology advantages contribute largely to foreign firms’ ownership advantages, which further enhance their asset power and overtake local firms in the market. Foreign-invested firms are more likely transfer the advanced and appropriate technology from the firm’s foreign parent (the multinational corporation), because the latter may be more willing to part with proprietary technology given its equity stake in the firm (Hu et. al, 2005).
Secondly, China has shown significant growth within the consumer electronics market and this can mostly be credited to the continual economic development of this densely populated country. However, the largest three producers in the Chinese market are the Japanese firms.11Although Japanese companies are inevitably important, Chinese
10 For example, over 2 million tones of ethylene capacity are set to go on-stream in 2005, and a further nearly 4 million tones are likely to be added in 2006 (as a comparison: German ethylene production in 2004:
5 million tones).
11 They are Sony Corporation, SANYO Electric Co., Ltd. Matsushita Electric Industrial Co., Ltd.
manufacturers also have significant market share.12 Government policies have a significant influence on the market as well.13 The reduced costs of production resulted form government’s intellectual property protection give domestic firms an advantage over foreign suppliers (Datamonitor, 2005). Thirdly, the Chinese textile industry is largely outsourced by MNCs. Possibly due to the fact that outsourcing or production sharing involves the putting out of work by firms such as branded manufacturers, wholesalers, retailers, and so on, in high-wage countries to independent firms in low- wage countries. As the costs of international transacting continue to decline, and as entrepreneurs in low-wage countries learn how to produce with global standards, textile industry in China ranks as No.1 in the world market (Scott, 2005). Fourthly, as Blackman and Wu (1998) applied data from an original survey of US private investors and official Chinese statistics, they assessed the volume and characteristics of FDI in China’s clean power sector. Their results imply that FDI is likely having a significant positive impact on the energy efficiency. Due to the reasons that China lacked the manufacturing wherewithal to supply the needed generating equipment14 and more importantly, China lacked the required financial resources.15 Generating these effects, the last hypothesis is:
H 4: Foreign owned firms perform better than domestic owned firms in the non-nature resource based and non-pollution intensive industries in China.
For different firm performance outcomes regarding different ownerships in China, predecessors have found some possible reasons to it. For instance, Holz and Lin (2001) has compiled data indicating that the lower profitability of DOFs can be fully explained by their higher tax rate and greater capital intensity, both of which reflect specific priorities and/or discrimination against DOFs. In addition, Wen (2005) found that the
12 For example, 80% of China’s DVD players are made by Jiangsu Shinco.
13 For example, the Enhanced Versatile Disc (EVD) standard was developed in China and adopted as the official standard for optical discs. This means that companies such as Jiangsu Shinco can produce EVD players and avoid paying royalties to DVD patent holders.
14 Its production capacity was estimated at between 9 and 12.5 GW per year, significantly lower than yearly expansion targets of 14.6}23.5 GW (Dorian, 1995; Murray et al., 1998).
15 Power authorities estimated that China would only be able to "nance 80% of the investment needed to meet its year 2000 capacity target with domestic resources (Shi, 1997). This implied that a total of 18 GW of new foreignfunded capacity would be needed between 1996 and 2000. Assuming average capacity costs of $US 600 to 800 per kW, the total capital required would be $US 11 to 14 billion.
unfair tax burden on DOFs is obvious. The taxes collected from DOFs were about three times as their profits, while the taxes collected from FOFs were about equal to or less than their profits. Possible explanations could be that, to encourage FDI inflows, the Chinese government has introduced various investment promotion policies and regulations. The incentives available include significant reductions in national and local income taxes, land fees, import and export duties, and priority treatment in obtaining basic infrastructure services towards foreign investors (OECD, 2000). Let along these policies, we want to know more factors that cause the differences in performance.
Analyses in the earlier researches were based on a simple comparison for the performance of domestic sample groups versus foreign sample groups without controlling for other relevant factors. FOFs may be attracted by industries that have above-average profitability levels; also FOFs may get clustered in industries that enjoy above-average performance relative to other industries. To control for the possibility of such factors in the estimates, it calls a number of variables that can positively or negatively affect firms in their ability to attain above-average levels of performance (Chhibber and Majundar, 1999). Besides, performance differences can emerge from either economic or organizational sources. Given ownership structures and industry contexts, firms make choices according to the various strategies they adopted. Industry related characteristics describe differences in industry performance. To analyze inter-industry differences in performance, a comprehensive framework developed by Caves and Barton (1990) 16 is supportive for this study. In particular, the major factors that help explain industry level performance variations deal with the effects of competitive conditions within an industry;
organizational factors; heterogeneity; government and regulatory policies (Hansen and Wernerfelt, 1989). By examining the industry conditions particular to China, I now introduce a number of variables that are derived from these five categories to explain performance.
Competitive conditions: including the export to sales ratio, imports to finished goods ratio and capital to output ratio. Since the former two ratios are not available on the China
16 The framework has been applied to explain industry level performance differences across in six countries.
Statistical Yearbook of NBS17, here I pick up capital related factor as the indicator. As enterprise that has higher working capitals is also likely to be more progressive and risk oriented, with relatively better performance patterns likely to be noted (Chhibber and Majundar, 1999). In addition, membership in WTO is assumed to exert pressures on Chinese firms to attain superior performance, because competitive intensity rises as a result of capital accumulation. So, Annual average balance of circulation funds (ABCF), which refers to the average value of all funds and capitals that can be used by enterprises during a reference period, is introduced as a control variable in the model to account for how competition affects performance.
Organizational factors: including the structure, systems, size and history of an enterprise.
Given that only size can be mathematically analyzed for the purpose of this research, I choose size as an indicator. Theoretically, the size of an enterprise can affect its performance in many ways. Key features of a large enterprise include its diverse capabilities, the ability to exploit economies of scale, and the formalization of procedures.
These characteristics, by making the implementation of operations more effective, can allow larger enterprise to generate larger returns on assets and sales. Alternatively, larger enterprise could be less efficient because of the loss of control by top managers over strategic and operational activities within the firm. Smaller enterprise, while unable to enjoy the advantages of economies of scale, is less hierarchical and bureaucratic and can be more flexible in adapting to situations where rapid decision making can allow itself to obtain larger than average profits. As a result, size can have both positive and negative consequences on performance (Chhibber and Majundar, 1999). For the purpose of industry level analysis, one bring Gross Industry Output Value(GIOV) as an indicator of size. It is the total volume of products sold or available for sale in value terms. It reflects the total achievements and overall scale of production for a given industry during one period. Gross industry output includes value of finished products, income from external processing, and value of change in semi-finished products at the end and at the beginning of the reference period. Since 1995, it was substituted by the gross industry output value by new method. Hence, an aggregate level of Gross Industry Output Value divided by the
17
total number of enterprise in that industry reflects the average size of per enterprise.
Specifically, the higher the gross industry output value, the lager the enterprise is.
Heterogeneity: measured by the amount of money devoted to advertising, marketing and distribution. As strategic choices affect the performance of enterprise within an industry and lead to the creation of heterogeneity within industries. Creditor’s equities (CE) are defined as investor’s ownership of net assets for the enterprise. It equals to the total amount of liabilities, which reflects the percentage of assets financed by creditors. A large literature has found creditor’s equities and entrepreneur performance to be positively related. The more equities the creditor have, the more expenditures that the firm is able to invest, for instance, marketing expenditures lead to more information about existing products; distribution expenditures widen the physical range of coverage and the variable proxies for geographic market heterogeneity. The amount of expenditures incurred on marketing and distribution activities as a ratio of sales can potentially enhance the per-unit revenue realization and lead to greater profitability (Chhibber and Majundar, 1999). The creditors’ equity ratio equals total liabilities divided by total assets.
This reflects the percentage of assets financed by creditors. When the total of creditor’
equity ratio is less than zero, that indicates the liability of the enterprise is smaller that its assets.
Government and regulatory policies: often considered as the tax rate. Knowing that the indirect taxes are major sources of revenue for government in some countries, enterprises act as tax collectors by recovering excise duties from the final customers and passing them on to the government authorities. The role of an enterprise as a tax collector can mute incentives to maximize profits because it operates more as an indirect revenue raiser for the government rather than as a business enterprise (Chhibber and Majundar, 1999).
Additionally, as being discussed in the literature review part, different industries are charged by different tax rates, since some are more pollutant than others; and different ownerships also enjoy different tax systems, either because of government’s monopoly or protectionism. So, Tax and Sales Extra Charges (TSEC) is taken to measure the performance effects that are resulted by government and regulatory policies. It refers to the amount of the value-added tax which should be paid by the enterprises during a
reference period. It is the annual accumulation of corresponding item in the “profit table”
of the accountant. For enterprises that do not follow the 2001 Enterprise Accounting Standards, the year-end accumulation of tax and extra charges from the sales of products is used as a substitute.
Once controlling for these influential factors in the estimates, the third specific research question asks:
3. What are the reasons that underpin the difference between domestic owned firms and foreign owned firm in terms of performance in China?
3. DATA AND METHODOLOGY
Having been discussed in the literature review part, some of the former studies on ownership and industry performance found insignificant outcomes. For instance, Wei and Varela (2002) tested the relation between state equity ownership and firm market performance for China’s newly privatized firms from 1994 to 1996. Results show that state ownership has a negative effect on firm value. The effect of international ownership is unpredictable and domestic institutional ownership does not appear to improve performance, possibly because the latter one lacks proper incentives to positively influence the firm’s management. Jefferson et. al (2005) examined differences in marginal factor productivity across ownership types, considers the impact of business cycles on the interpretation of productivity trends, and documents a statistical relationship between the profitability of state enterprises, relative productivity performance of state firms, and entry of new firms outside the state sector. Research on industry productivity began with studies of SOEs’ value added, capital, and labor. Chhibber and Majundar (1999) studied the influence of foreign ownership on the performance of firms operating in India. Foreign ownership is categorized according to the control exercisable at different levels of ownership. The results show that, after controlling for a variety of firm and environment-specific factors, only when property rights devolve to foreign owners, at ownership levels providing unambiguous control at 51 percent, do firms in which there is foreign ownership display relatively superior performance. Boardman and Vining (1989) analyzed the effect of ownership in a competitive environment by controlling for relevant factors. They found the most convincing test of the property rights theory as basing on a comparison of industry state-owned enterprises (SOEs), mixed enterprises (MES) and private corporations (PCs). Table 2 summarizes their methodologies and results.
Table. 2 Previous studies and their results
Effects of ownership on industry performance
Study Country Sample Data type Statistic techniques
Measurements of performance
Results
Wei and Varela,
2002
China
652 firms listed on the Shanghai Stock Exchange
Firm level data
Ordinary least squares
(OLS) regression
Tobin’s Q and
MSR Insignificant
Jefferson
et al, 2005 China Industries listed on NBS
Aggregate industry level data
Simple regression
model
Marginal factor
productivity Insignificant
Chhibber and Majundar,
1999
India
1,000 Indian firms listed on
the Bombay Stock Exchange
Firm level data
Ordinary least squares
(OLS) regression
Return on sales (ROS) and Return on assets (ROA),
Insignificant
Boardman and Vining,
1989
Canada, Germany,
Britain, Japan and other
countries
500 largest manufacturing
and mining corporations
National wide industry level data
Ordinary least squares
(OLS) regression
Return on equity (ROE), Return on assets (ROA),
Return on sales (ROS), and Net income (NI)
Significant
Demsetz and Lehn,
1985
U.S.
511 firms from major sectors of
the U.S. economy,
including regulated utilities and
financial institutions;
406 firms of manufacturing
and mining
Firm level data
Ordinary least squares
(OLS) regression
Profit rate Insignificant
As Lester (1996) states: “It is not enough that a research study involve relevant, interesting questions and be carefully conceptualized, designed, and reported. What’s more, the conduct of the study itself must include the effective application of appropriate data collection, analysis stage, and interpretation techniques. The principles for designing instruments, reducing data, and selecting samples have been developed within various disciplines to guide researchers in carrying out their studies. Although these principles should not be followed strictly, competent researchers always consider them in order to ensure that every aspect of their studies is appropriately and carefully carried out.”
Acknowledged by this argument, the advantages and disadvantages of each methodology are summed up as below.
Table. 3 Advantages and disadvantages of previous studies
Study Advantages Disadvantages
Wei and Varela, 2002
• High reliability of the data from the stock exchange information.
• The panel of the sample as two years is inconvincible.
• Firm level data gives insignificant results
Jefferson et al, 2000
• Simple regression model is easy to taken and re-checked.
• Testing marginal factor productivity18 as the dependent variable gives answers under a assumption of profit-maximizing theory. It is too microeconomic concerned, may not be applicable for a country specificmacro case.
• Firm level data gives insignificant results Chhibber and
Majundar, 1999
• High reliability of the data from the stock exchange information.
• Relatively large sample size.
• Firm level data gives insignificant results
Boardman and Vining,
1989
• Cross- country comparison
• Different variables to measure performance give robust results
• Large sample size
• National wide industry level data gives significant results
• The sample countries picked up share a lot of similarities, especially in terms of the control variables. Therefore, the result of the cross- country compassion is not widely applicable
Demsetz and Lehn, 1985
• Large sample size
• Types of firms are not identical, including regulated utilities , financial institutions and manufacturing firms
• Firm level data gives insignificant results
As can be seem from the various previous methodologies, samples are either picked up from the stock exchange market, the national bureau of statistics (NBS) or from the company report. Among all the results, only the national wide industry level data studied by Boardman and Vining (1989) gives significant results. Most of the researchers adopted the Ordinary least squares (OLS) regression as their statistic technique. And Demsetz and Lehn (1985) indicated profit rate as their performance measurement variable. Motivated by these preceding researchers, this paper is going to collect data from China's National Bureau of Statistics (NBS), analyze specific industry level data, build up an ordinary
18Marginal-productivity theory indicates that the demand for a factor of production is based on the marginal product of the factor. In particular, a firm is generally willing to pay a higher price for an input that is more productive and contributes more to output. The demand for an input is thus best termed a derived demand. See, http://www.amosweb.com/cgi-
bin/awb_nav.pl?s=wpd&c=dsp&k=marginal+productivity+theory
least squares (OLS) regression model with a few control variables added, and evaluate performance by profit. In this part, an exhaustive description of the chosen methodology for the paper will be illustrated. It will indicate how the methodology works as well as provide information on the source of data.
Data Collection
According to Jefferson et. al (2005), the sources of data for studies on an industry level are usually the national statistical data. China's National Bureau of Statistics (NBS and formerly the State Statistics Bureau) gathers data for different ownerships by their status of registration. There are, for instance, state owned firms (SOEs), collective firms (COEs, including township and village enterprises), and firms organized under foreign ownerships (FOFs). Table 4 describes these classifications and explains how they are aggregated into two components for the purpose of this study: domestic owned enterprises (DOF) and foreign owned enterprises (FOF).
Table. 4 Classification of enterprises by status of registration
Domestic Owned Enterprises (DOF) Foreign Owned Enterprises (FOF)
• State-owned enterprises
• Collective-owned enterprises
• Cooperative enterprise
• Joint ownership enterprises
• State joint ownership enterprises
• Collective joint ownership enterprises
• Joint state-collective enterprises
• Other joint ownership enterprises
• Limited liability enterprises
• State sole funded corporations
• Other limited liability corporations
• Share-holding corporations limited
• Private enterprises
• Private-funded enterprises
• Private partnership enterprises
• Private share-holding corporations Ltd.
• Other enterprises
• Enterprises with funds from Hong Kong, Macao and Taiwan
• Joint-ventures enterprises
• Cooperative enterprises
• Enterprises with sole investment
• Share-holding corporations Ltd.
with investment
• Foreign funded enterprises
• Joint-ventures enterprises
• Cooperative enterprises
• Enterprises with sole foreign funds
• Share-holding corporations Ltd.
with foreign investment
Source: China Statistical Yearbook (2006)
One limitation of Jefferson et. al (2000) is that their data include all forms of ownerships, but not all forms of industries, as they are confined to enterprises with independent
accounting methods.19 This research is going to recover this gap, all industry sectors, according to NBS, will be put into consideration. Thus, our data replicates the basic conditions of all industry sectors, including main industry economic indicators of 31 provinces, autonomous regions and municipalities. 20 Moreover, industry statistics cover all industry enterprises within the Chinese territory but excluding Hong Kong, Macao and Taiwan (NBS, 2006).
However, there are two shortcomings of the data source from NBS, the first is that the statistical period for segmenting ownerships is different, due to the different government policies, the development speed and the change of status. For instance, official figures partition the FOF sector into classifications changed substantially in 1993, and there are no indicators of FOF by industry brunches until 1999. The second weakness is the reliability of the data, since the industry enterprises statistics in NBS are collected from Year Report of Industry Statistics, which are provided by statistical bureaus of each regions through Comprehensive Reports to NBS. Although statistical agencies at various levels do arrange and collect data through comprehensive reporting system basing of real situation in local areas (NBS, 2006), as Jefferson et. al (2005) exampled, year 1996 data encompass 74 percent of gross industry output, but less than one-half million of China's nearly 8 million industry enterprises. It is probably caused by the changing number of independent accounting units within specific each ownership categories, sometimes substantially, from year to year. Meanwhile, focusing on independent accounting units reduces the concerns about data quality. For these reasons, Chinese and international researchers have commented extensively on upward bias in the output data for state and especially for collective industry (Woo et al, 1993; Jefferson, Rawski, and Zheng, 1996;
Lardy, 1999). Restrictions on the accessibility of the data and the awareness of reducing the comprehensiveness of coverage, call this analysis to focus more on the available data segments. Therefore, seven years panel data, covering sources from year 1999 to 2005 from NBS will be treated specifically.
19 The data exclude industries operated by village-level enterprises, individual proprietorships (employing seven or fewer workers), and establishments whose results are incorporated into the accounts of other units.
20 Data by industries are based on the 2002 National Industry Classification of all Economic Activities.
Reviewing Table 1 for the industry segmentations, there are 37 manufacturing industries in China each year. As this research contains two types of ownership, 37 times 2 implies that there are 74 sections by both ownerships. Since the panel contains 7 years, the total number of observations for this research is, thus, 74 times 7 gives 518 unites.
Methodology
In this study, a regression model will be built to test the positive and negative relation between ownership and performance, plus its degree of influence. Followed by the ideas of Chhibber and Majundar (1999) and Jefferson et. al (2005), the method of the regression is Panel Least Squares. The regression takes the form of:
P it =α+ β1 O + β2 TOI + β3 ABCF it + β4 GIOV it+ β5 CE it + β6 TSEC it + e (i) where P is performance, O stands for DOF or FOF; TOI represents different industry types as shown in table 1; ABCF represents the annual Average Balance of Circulating Funds by a specific industry; GIOV stands for the Gross Industry Output Value; CE means Creditor’s Equities and TSEC signifies Tax and Sales Extra Charges; and eis an error term. Besides, i (i =1,2,3,4) is the industry type corresponding to the different industry types as shown in table 1; and t (t =1999-2005) indicates the time.
Dependent variable
The dependent variable P is the performance of the each industry type by its ownership, it is measured by the total profits. According to NBS, the unit of analysis is 100 million Chinese Yuans, which is around 13.21 million US Dollars.21
Independent variables – O, TOI
Ownership(O), is measured by dummy variables as:
O=1, if the firms are domestic owned firms (DOFs) O=0, if the firms are foreign owned firms (FOFs)
Type of industry(TOI), is a nominal variable which takes the value as:
TOI =1, if it is type I industry (Nature resource based an pollution intensive industry) TOI =2, if it is type II industry (Non-nature resource based an pollution intensive industry)
21 This is calculated by the currency exchange rate between Chinese Yuan and US Dollar on the 13th of July,
