This study investigates the interactions between provincial characteristics and foreign-specific agglomerations on Taiwanese investors' location choice in China. Using firm-level data, we find that nationality agglomeration and Asian agglomeration have non-negligible impacts on these investors. Furthermore, we find that their location choice follows a sequential selection process. These findings suggest that a region-wide development strategy is a more effective means of attracting these investors than province-specific fiscal concessions and preferential treatment.

Attracting foreign direct investment (FDI) remains a key developmental strategy that is prescribed by policy-makers in many developing economies. It rests on the premise that FDI accelerates economic growth in the host country through both tangible and intangible channels. In the extant literature, China’s dominance in the world’s vast FDI scene has captured many researchers’ attention. However, the majority of the existing studies have focused on either the national or provincial level, leaving the determinants of FDI and its effects on the host city largely under-explored. An in-depth understanding of the reasons behind the mixed experience of Chinese cities in hosting FDI provides invaluable input for local officials. With these points in mind, this thesis is partly designed to bridge the gap in our current understanding by examining FDI using city-level data from China, and, partly, to provide the Chinese policymakers with an insight into the interactions between FDI and the host city.

A comprehensive panel dataset of 287 prefectural cities in China for the 2001–2012 period forms the basis of this thesis. Findings from this thesis complement earlier results and offer important lessons for local development. From a methodological perspective, this thesis addresses aggregation bias by considering various subsamples based on geographical topology and city-specific characteristics. Furthermore, it controls potential endogeneity bias by applying panel instrumental variable (IV) or generalised method-of-moment (GMM) estimators.

This thesis is organised in a paper-based format and presented in three parts. Part I provides essential background information that links various facets of this thesis together. Part II presents the results of the empirical analyses in five distinct, but related, papers that are delineated by specific research questions and objectives. Part III discusses the main findings and their policy implications. This final part also lists the limitations of the current research and offers directions for future research.

The first two papers focus on the city-level determinants of FDI. Specifically, the first paper investigates the role of soft infrastructure as an attractor of FDI. After capturing the notion of soft infrastructure by the local banking sector (LBS), it finds a positive association between LBS development and FDI across Chinese cities. This finding suggests that local governments need to devote their limited resources to improving soft infrastructure, like access to financial services. Meanwhile, the second paper addresses the effect of capital-city status in attracting FDI. It finds that, all things being equal, capital cities manage to host more FDI compared to cities without capital-city status. This finding implies that Chinese policy-makers need to focus their attention on non-capital cities in order to achieve a balanced growth path within the province.

The remaining three papers apply GMM estimator to study the interactions between FDI and its host city in China. Paper three investigates whether FDI complements or substitutes for domestic investment (DI) in the Chinese cities. The analysis shows that a complementary FDI– DI nexus only exists in the eastern and central cities. This heterogeneous relationship is further supported by the results in paper four, which attributes the positive FDI–growth nexus in the eastern cities to soft infrastructure, such as local financial development and human capital, and in the inland cities to hard infrastructure, like transportation networks. Meanwhile, paper five finds a unidirectional causality running from FDI to growth in the eastern cities, but a bidirectional causality in the central cities. In general, these findings attest to the argument that the interactions between FDI and the host economy crucially depend on initial economic conditions and the local institutional environment. Without taking these city-specific characteristics into consideration, any attempt to attract and internalise the benefits associated with FDI is likely to be futile.

Overall, this thesis provides two major contributionsto the FDI literature. First, it demonstrates that soft infrastructure has played an increasingly important role in influencing the location choices of foreign investors in China. Second, it highlights the need to study the determinants and effects of FDI on a smaller geographical scale, particularly in a large country such as China.

This paper seeks to explain uneven distribution of Taiwanese direct investment in China. Using firm-level data, we find that such a pattern can be attributed to different characteristics of key economic regions. Furthermore, we find evidence that the probability of a province being selected by a Taiwanese investor increases with the extent of industrial linkages and labour-market pooling. Importantly, we find evidence that the structure of the Retail and Service industry's location choice is markedly different to its Manufacturing counterpart.