Comments on Revisiting Complementarity between Japanese FDI and the Import of Intermediate Goods: Agglomeration Effects and Parent-firm Heterogeneity

Abstract

By tracking the detailed product-level trade transactions for 929 Japanese affiliates in China, Ito, Matsuura, and Yang contribute to the classic literature that investigates the complementarity versus sub-stitutability of trade and FDI. To advance this question, the paper studies the duration of intermediate input imports by these Chinese affiliates between 2000 and 2006. Because the authors have taken on the arduous task of creating a data set that combines four different sources, they illuminate a number of interesting relationships that are generally obscured by firm efforts to hide their proprietary actions from view. The key question posed by this paper is: How is the duration of intermediate input import by Chinese affiliates shaped by product, firm-level, and national considerations? Thanks to the paper’s careful data effort, much of the paper’s novelty lies in its ability to relate this question to the Japanese parent firms’ characteristics and global footprints. Most notably, the paper shows that the duration of affiliate import of intermediate inputs was shorter when the Japanese parent firm was larger (as measured by employment), had a larger global network of affiliates, or had longer experience operating as a multinational. Although this inquiry piques the reader’s interest in this question, however, the generaliz-ability of the results are called into question by the uniqueness of the period of estimation, 2000–06, which lies solidly in the interval of China’s accession to the WTO and coincides with the period of rapid hollowing out of Japan’s manufacturing sector. 1 In this context, for Japanese industries relocating to China, the main question was sequencing. Firms could move everything at once, or start assembly early on, using imported inputs that were later replaced by new suppliers. To the extent that Japan’s affiliates in China were initiating operations that were replacing their predecessor operations in Japan, the duration analysis reveals the product trades that were most quickly relocated. Intuitively, the paper shows