Abstract: For many years, despite the lack of sufficient theoretical basis and empirical support, the idea that high value-added should be the direction of manufacturing industry upgrading has long occupied the mainstream position in the domestic field of industrial structure research, and it is also an important proposition of government′s industrial policy. Based on the cross-sectional data of 72 countries in 2014 and the panel data of 58 countries from 1997 to 2014, this paper makes a number of empirical tests on the existence of high value-added trends in the process of economic growth and manufacturing industry development, and all the corresponding analysis results do not support this statement. Furthermore, based on the sub-industry panel data of 19 OECD countries from 1995 to 2013, this paper also shows that, at least at the level of the main classes, there is no evidence to confirm that an industry is high (low) value-added in various or most countries, the higher (lower) value-added rate is only the inherent property of a few industries, and economic growth in high value-added industries is not faster than the others. Therefore, this paper holds that although economic growth and industrial structure transformation have many characteristics and even rules to follow, the profit-driven enterprise may not create a so-called high value-added trend, and increasing the value-added rate should not be one of the specific objectives of the government′s industrial upgrading policy.

Key words: high value-added, industrial upgrading, manufacturing industry, value-added rate, trend