关键词: 外商直接投资, 动态Malmquist模型, 权重多维尺度分析, 全要素生产率, 区位选择

Abstract: With the transition of new normal economy in China, the economic growth performance becomes more importance. According to the 2017 World Investment Report, the global foreign direct investment (FDI) inflow was USD1.52 trillion in 2015, with an annual decreasing rate of 16%. This means, the FDI of developing economies is decreasing because of a decrease in prices of staple commodities, the devaluation of currency, and an increase in geopolitical risks. Nevertheless, the Chinese economy continues to grow despite the downtrend. With an annual growth of USD 131 billion, China ranks second in FDI inflows, after the United States. Thus, FDI is essential for economic development in China. During the planned to market economy transformation, the Chinese economy was promoted with a regional cascade structure. Subsequently, openness in China was bolstered using the same structure. Consequently, the policy supports and complementary conditions of FDI importation vary among the provinces. Therefore, the FDI importation performance exhibits extreme variations among the provinces. The FDI in China is disproportionately distributed. To understand the influence of intertemporal effects of dynamic factors on FDI importation sustainability, this paper examined FDI importation in Chinese provinces to investigate the differences in importation performance in a single-country setting, where the effects of single-period support conditions and dynamic factors, such as energy consumption and physical capital stock, are adequately controlled.To identify the differences in FDI importation performance among provinces, this paper selected 30 provinces in China, excluding Tibet, as DMUs. Previous studies have indicated that the influences of total factor productivity focus on physical capital stock, human capital stock and output value, based on economic growth theory. In addition, hysteresis of physical capital stock exists in economic growth and technical progress. In other words, the sustainability of total factor productivity is not only represented by the congruence of influence factors but also reflected in the sustainability of physical capital stock. From the perspective of the supply-side structural reform, in order to reveal the FDI regional disparity, this paper constructed a dynamic Malmquist model to evaluate the dynamic total factor productivity of FDI distribution by considering the inter-temporal effect.Based on provincial panel data of China, this model used human capital, material capital, energy consumption, and openness as inputs and FDI performance as outputs. In addition to the input and output factors, this paper defined physical capital stock as a dynamic factor considering the intertemporal effect of hysteresis. Furthermore, multidimensional scaling is conducted to demonstrate the emphasis of FDI input efficiency, named soft power and hard power, respectively. Moreover, the average data is used to eliminate the influence of population differences among provinces. Thus, the average physical capital stock is defined as a dynamic factor. The average FDI is defined as output. The average human capital stock, average energy consumption, and export rate are defined as input factors. This paper derived all data from official statistics, such as those in the China Statistical Yearbook, China City Statistical Yearbook, and China Energy Statistical Yearbook.This study examined the different dimensions of the FDI importation performance among Chinese provinces. For this purpose, an expanded dynamic Malmquist model is built with the inclusion of additional inter-temporal effects. This paper not only evaluates the FDI growth performance of China′s provinces from the perspective of energy conservation, but also makes an in-depth analysis of the driving force of China′s regional FDI development from the perspective of differentiation. In the way, the energy-saving task under the premise of stable growth can be realized. As expected, the DTC is the major bottleneck in FDI importation performance. This finding is consistent with those of previous studies on economic sustainability. Therefore, this study is a valuable complementary research that empirically clarifies direct effects of FDI importation. Furthermore, this study establishes a critical bridge between FDI importation and sustainability expectation, although many previous studies have frequently linked FDI importation to its extrinsic outcomes, such as profits or financial performance, without considering the inter-temporal effect. The study showed smart power playing the role of the bottleneck issues of FDI regional disparity. Long-term of policy dividend and local conditions are important issues for choices of FDI investment distributions. The empirical results provide preliminary evidence of FDI importation performance bottlenecks, which constitute a nascent yet emerging field that has considerable potential for future research. Based on the results, this paper proposed the following suggestions as a reference.First is to consider local characteristics. From the perspective of FDI total factor productivity of 30 provinces (except Tibet) in China, it is significant for the imbalance of FDI location choice. The constraint bottlenecks of different investment locations are diversity. Therefore, "Adapting measures considering local conditions" is the primary strategy in FDI investment location selection. On the one hand, formulating FDI introduction strategies, every province should not only integrate the national macroeconomic situation, but also combine the actual development in their own region. Besides that, it is important to consider the economic development experience of neighboring regions. On the other hand, making investment location choice, investors need to understand the characteristics of various factor inputs of FDI and local policies. And then, they can match the investment location with their own industrial attributes. For making the collaborative innovation, it is very important to fully release the vitality of production factors.Second, compared with the stability of the central region, the eastern region tends to transform to smart power, while the western region relies on policy dividends. This can be verified from the policy dividend of the "western development". However, it came to the end of the second dividend in 2013. "One Belt And One Road" national strategy is the opportunity of the third policy dividend in the western region. The key for western provinces is to create new opportunities for development and realize complementary advantages. So, on the one hand, it is necessary to consider the characteristics of inter-temporal factors. On the other hand, the combination with the emphasis on the development dimension of FDI investment can lead the orderly free flow of economic factors, efficient allocation of resources and deep integration of the market.Third, the transformation to smart power is the new direction of FDI location selection with the development trend of supply-side structural reform. The empirical analysis shows that the focus of FDI location choice has shifted from the comparison of soft power to smart power. For example, the advantages of FDI location selection in Shanghai and Tianjin have been transformed into smart power. Liaoning is on the rise in the process of smart power transformation, while Guizhou Province is still on the trend of disadvantage in FDI location selection. The new model of FDI smart power transformation needs attention.

Key words: foreign direct investment, dynamic Malmquist, multidimensional scaling, total factor productivity, location choice