«The Effect of Technology Spillover from FDI: An Empirical analysis of Chinese Manufacturing Industries Li Li *Faculty of Economic Sciences, Hiroshima ...»
The Effect of Technology Spillover from FDI:
An Empirical analysis of Chinese Manufacturing Industries
*Faculty of Economic Sciences, Hiroshima Shudo University, Japan.
This paper examines the effects of technology spillover from FDI based on panel data
analysis of Chinese large-scale domestic manufacturing industries. We use the panel
data of 27 Chinese manufacturing industries from 2001 to 2007. We estimate technology spillover effects by two steps. First, we estimate the total factor productivity of domestic firms’ and foreign firms’ by Cobb-Douglas Production function. Second, we extend the previous studies to examine whether foreign firms give the effect of technology spillover to the Chinese domestic firms, and in this paper, Chinese 27 industries are classified into four different subclasses according to the levels of technology -- highest, higher, lower and lowest level. Then we compared the effects of technology spillover in different industries’ technology level. The result indicates that foreign firms have positive effect of technology spillover on Chinese large-scale domestic firms, and these effects of technology spillover is seen to be centered on highest, higher and lower technology levels of industries.
Keywords: FDI; TFP; Technology spillover.
1. Introduction Although in endogenous economic growth theory, technical innovation is the source of economic growth that accruing on the research and development voluntarily (R&D), in recent years, Asian developing countries have achieved a high economic growth rate, not through making R&D voluntarily, but through foreign capital firm’s entry, so they studied and copied foreign firm’s technology and know-how of management, and China was among these countries. China have attained a high growth rate and in the limelight, because Chinese government had successfully Promoting foreign direction attracted foreign direct investment (FDI) and created exportation.
investment (FDI) has always been a primary concern for economic growth, especially in developing countries. The Chinese government has encouraged FDI in order to prop up backward industries since 1978. After 1992, Deng Xiaoping's southern tour of Shenzhen encouraged further and much more massive wave of foreign direct investment, which contributed towards acceleration in GDP growth of China. Foreign Direct Investment (FDI) has exceeded about 60% in Chinese manufacturing industries since 1996, although that ratio has been downed to about 50% during the1997-1998 Asian financial Crisis. In 2001 just as China entered WTO, with the reform of regulations concerning trade, Chinese government made various improvements in environmental of foreign direct investment. Chinese manufacturing industries has increased up to about 70% in
2004. A few years later, because the tax of domestic firm was higher than foreign firm since the Chinese economic reform, loud cries of discontent were being voiced among the domestic firms, in order to relax the unfairness, Chinese government made reform of the same income tax rate for the domestic and foreign firms in 2006, so FDI inflows down to 60%. After that, the global financial crisis of 2008 forced FDI inflow to down to 50% in the Chinese manufacturing industries. However, the role of FDI has not been disregarded in the Chinese manufacturing industries growth. So, a lot of studies have taken up manufacturing industries in Chinese empirical study field.
Generally speaking, technology has a spread property. FDI not only brings a large of capital but also bring advanced technology, so FDI will give investment receiving countries economic effect1. Of course, foreign firms don’t expect competitor-Chinese domestic firm to improve their technology level, but Chinese domestic firms can benefit from technology spillovers from foreign entrants. Because foreign firms are not able to extract the full value of these gains, they are often called involuntary technology transfer or spillover effect by Kokko (1994).
Because of domestic firms’ technological gain, technology spillover effect from FDI generally results two channels. First, there are a number of channels through which FDI affects productivity of domestic firms. We suppose that foreign firms have a higher technology than Chinese domestic firms in the same industry, these foreign firms direct establish their subsidiary company in China, or contract with Chinese domestic firms to set up Cooperation or Joint-venture enterprises. In this paper, three foreign fund of enterprise referred above --enterprises with sole funds, cooperation enterprises, and joint-venture enterprise, we call them foreign-affiliated firm. Chinese domestic firms provide foreign firms with some materials, or domestic firm improves its productivity.
by simply observing nearby foreign firms and copying their technology, this case demonstration effect will take place. Second, another type of spillovers is through competition between foreign firms and domestic firms. The competition effect, unlike demonstration effect which is presumably positive, can be either positive or negative.
FDI may toughen the competition faced by domestic firms, thereby forcing them to become more competitive. On the other hand, increase in competition with inward FDI can also reduce productivity of domestically owned firms, particularly in the short run.
If imperfectly competitive firms have to incur fixed costs of production, a foreign firm with lower marginal costs will have an incentive to increase production relative to its domestic competitor. In this environment of market, entering foreign firms producing for the domestic market can steal demand from domestic firms, forcing them to reduce 1 For example, see Fan Janting (2004) “Industrial Development and International Division of China: FDI toward China and Technological Transfer”, P33(Japanese).
2. Literature Review As globalization proceeds, the study is being researched in effect of technology spillover. Most of studies used a Cobb-Douglas production function of Feder (1982), which we also use, adding foreign capital which comes from FDI to domestic firms.
Feder (1982) provide a new econometric framework for analyzing the spillover
share and β is capital share. Feder (1982) divided economics into foreign and domestic two sectors and added foreign capital into Cobb-Douglas production function as a factor of production.
where Y, K, L denote output, fixed capital and labor, KH is fixed capital of domestic firms, KF is fixed capital of foreign firms in China, LH is total labor of domestic firms and foreign firms in China, α is output elasticity of labor, β is output elasticity of domestic fixed capital. γ is output elasticity of foreign capital. Ifγis significantly positive, it is concluded that the factor enhances spillover. it restricts spillover if γ is significantly negative.
The technological spillover effect of FDI means that the technology is embodied in capital in Eq. (2.3). The results showed that domestic firms benefit significantly from foreign firms.
In the early 1990s, the Chinese surge in FDI has attracted the attention of economists. But in recent years, a lot of empirical studies have tended to take analysis of technology spillover effect from FDI in Chinese industries. Most of studies use time series and their results indicate that FDI has positive effect of technology spillover on Chinese domestic firms. The others studies use total factor productivity based on Cobb-Douglas production function (Todo, 2008; Li Xing, 2008).
Li Xing (2008) in Chapter 4 estimate technological spillover of FDI in china, using panel data,
where Yit denotes value-added for industry i in year t, Ait is total factor productivity for industry i in year t, and Lit, Kit are labor and capital for industry i in year t, Bit is the others affect factor except FDI for industry i in year t. ηis technology Spillover effects factor of FDI. Shareit is rate of foreign capital to total capital of Chinese manufacturing industries for industry i in year t.
Substitute Eq. (2.4) For Eq.(2.5)and taking natural log, can be written as following:
αis output elasticity of capital and β is output elasticity of domestic fixed labor. η is technology elasticity of foreign share.μit is error for industry i in year t.
If ηis significantly positive, it is concluded that the factor enhances spillover, and it restricts spillover if ηis significantly negative. The result showed that domestic firms' productivity or technological growths by foreign firms enter.
BaoQun and Lai Mingyong(2003) estimate the effects of technology spillover of
Above studies are analyses of macro-level, but not analysis of micro-level industry.
In this paper, we will examine the effects of technology spillover from FDI based on panel data analysis of Chinese large-scale domestic firms2, and we use the panel data of 27 Chinese manufacturing industries from 2001 to 2007. We estimate technology spillover effects by two steps. First, we estimate the total factor productivity of domestic firms and foreign firms by Cobb- Douglas Production function. Second, we extend the previous studies to examine whether foreign firms give the effects of technology spillover to the Chinese domestic firms in micro-level industry, and here Chinese 27 industries are classified into four different subclasses according to the levels of technology. So we can compare these effects of technology spillover are seen to be centered on which technology levels of industries by pooled least squares.
3. Empirical Methodology In this section, we estimate technological spillover effects by two steps. First, to estimate technology level, we borrow Cobb-Douglas production function. After that, we extend the previous studies to examine the effects of technology spillover of FDI on large-scale domestic manufacturing industries in China.
3.1 Estimating for technology level The first, we consider research methodology to estimate technology level. As is well known, the Cobb-Douglas production function is used often when economists estimate
productivity. We also borrow Cobb-Douglas production function as following:
Where Y is output, Y, is dependent upon two input factors-Physical capital, K, Labor, L, A is technology level, denotes the others input factors expect labor and capital, the others studies name it total factor productivity,α is labor share and β is capital share, ε is error.
For Eq. (3.2) using panel data with industry i and year t, and giving natural log, it
can be turned as following:
Where i, is industry, and t is year, Yit, denotes value-added for industry i in year t, Ait, is technology level for industry i in year t, and Lit, Kit are labor and capital for industry i in year t, εit, is error for industry i in yearｔ. α,is output elasticity of labor, and β, is output elasticity of fixed capital.
When the constant returns to scale prevail or α+β=1, Eq. (3.3) can be expressed as:
(3.4) Using Panel Ordinary Least Square (OLS) estimator on Eq. (3.4), βcan be estimated.
Because we supposed α =1-β, so we also know α, Eq. (3.2) with industry i and year t,
can be written as following:
Based on Eq. (3.5), using the data of β,α,and the origin data of Yit, Lit, Kit, we can estimate the technology level of foreign firms (AFDIit) and technology level of Chinese large-scale firms (Ait).
3.2. Estimating for technological spillover effects As the second step, we consider whether these factors have effects on technological spillover. In this paper, we consider affecting factors as following, Ait, is dependent upon following three affecting factors.
Our key assumption is that the foreign firms and domestic firms have the competitive effects, add the foreign capital (FYit-1) exactly the same as the previous studies, is expressed as Eq.(3.7), and we extend the previous studies and take up technology level of foreign firms (AFDIi,t) too. For domestic firms, we can consider the relationship between technology level in year t (Ait) and its t-1 (Ai,t-1), We provide a new
econometric framework which is expressed as follows:
where Ait is total factor productivity for industry i in year t, just physical capital and labor are impossible expressed all the relation of production, as the others affect factors, it is not only technological growth by the change of the times, but also adjustment result about peculiar technology to industry, for example-- The change of factor price, or not able to observation by data. Technology innovation such as according to the firm’s entry and leave in the same industry, an efficient distribute effect of production factor. Bit is the other factors affecting technology level of domestic firms. AFDIit is technology level of foreign firms for industry i in year t, FYit-1 is the rate of foreign capital to output of domestic firms in industry level in year t-1. Ai.t-1 is total factor productivity in year t-1, μit
is error for industry i in year t, which be expressed as following:
where αis a constant, αi is technology level of domestic firms elasticity of technology level of foreign firms in 27 industries level, here we call effect of technology spillover from FDI. β is technology level of domestic firms elasticity of FY. γ is technology level of domestic firms in year t elasticity of its in year t-1. Considering time lag, we give time-lag 1 to A and FY as independent variable.
Using Pooled Least Squares estimator, we can estimate α i, if coefficientαi is significantly positive, it is concluded that the factor enhances spillover effects, and it restricts spillover if αi is significantly negative.
4. Data Set