1.Introduction

As we know, China's emergence was mostly led by export.Since the reform of opening up, China has enhanced export through various channels like reducing cost of production, shifting structure of export, practicing policies of subvention in order to develop key sectors.We has already studied China's cost advantages（SU 2011）thus in this paper we would like to talk about the structural change.When measuring specialization by export share, China was indeed more specialized in sophisticated sectors.For instance, export of electronics at high-tech level leapt from 3% of world trade in 1990 to about 1/4 in 2010. That of electric sector also rose above 20% recently.

However with the deepen division of labour and global integration, an exported product must more or less contain imported inputs from abroad.According to the theory of comparative advantage of Ricardo, each partner countries gain from trade.Question is how distribute this gain between different partners.Greg Linden（2007）gives an example of“IPod”30GB of 5e generation.Every 300$ IPod sold in US takes 150$ factory cost, of which China holds 4$ and US 14$.China custom statistics define an IPod export as 150$ surplus for China, i.e.150$ deficit for US, but actually US gain much more than the former（at least 14$-4$=10$）.Measurement by pure export statistics will therefore conduct bias conclusions.

Hummels, Ishii, Yi（HIY 1999）propose the concept of Vertical Specialization and measurement by Input Output Table, which permits to evaluate changes in the“nature”of international trade according to the stage of production.Chen（2001）analyses China-US trade with the same method.He notes that although china experienced a rapidly increasing export, its local content did not rise as much as US.This is why China gained less than US.According to the study of Koopman et al.（2008）, the local content of Chinese export for high-skilled sectors was lower than that for low-skilled sectors.Chen（2008）pointes that the export of sophisticated sectors cannot elevate china's domestic value added and the later decreased during 1995-2002.This paper aims at analysing what led China's structural changes.If they were led by domestic production, China gained more from structural transformation.If they relied on increasing imports of inputs, China gained less from a unit of export.

We utilize Vertical Specialization and Input Output（I/O）Table to assess china's local content of export over 1995-2007.It has three contributions:1）existing articles mostly apply I/O Table of 2002.We extend the period to 2005 and 2007.2）We compute vertical specialization share, also named imported inputs in china's production for export of 135 sectors, which is the most disaggregated among recent researches.Besides, we match the classification of China with that of OECD, reconstitute 135 sectors'I/O table into 48 sectors'table and compare our findings with OECD's calculation for confirming our robustness.3）A large number of researchers argue that Chinese export depends more on inputs' import than before but few of them prove it.We use a macroeconomic equation to estimate linear correlation between export and foreign content then obtain a further result of dependence variation.The study is organized as follows.Section 2 presents China's structural changes of export during 1970-2009 by stage of production, technological level and sector.Section 3 analyses the foreign value added, also the vertical specialization share of export between 1995 and 2007.It firstly introduces the theory and the selected methodology; secondly explains data sources and comparison; finally combines I/O tables with trade data and illustrates preliminary results.Section 4 provides empirical studies to confirm the obtained inference.Section 5 gives conclusions and future directions of research.