738 D. Lo and. h chan 100.00 Industrial 50.00 40.00 10.00 Figure 1. Output composition of Chinas electronics industry, 1983-95(sectoral shares at 1980 constant-price gross output value). Sources: ZDGN-D and ZdGn (various years): ZDB (5 March 1996 p 4) and Jiang(1993)are among the very few studies in the literature that deal with this topic, both covering mainly the second half of the 1980s. The results which they have obtained do not agree with each other, largely due to their different estimation methodologies as well as different ways of aggregating data in the attempts to achieve correspondence between trade and production categories. A further shortcoming that might serIous sly limit the usefulness of such studies concerns the comparability of the capital stocks of different industries. It is a salient feature of China's manufacturing sector that some industries(e. g, electronics) are much newer than others. As a result, the value figures of capital might obscure the fact that capital productivity differs widely across industries, especially because of problems of pricing and depreciation that are inherited from the pre-reform era. In any event, it can be seen from Table 3 that the capital-labour ratio of the mechanical and electronics industry and that of he electronics industry alone have both outstripped the manufacturing sector as a whole, and, in the latter case, the gap has widened over time able 4 presents an alternative measure that is simple yet effective in indicating the relative capital and technology intensity in production: the per worker net output value of the mechanical and electronics industry, the electronics industry, and the manufacturing sector as a whole. For the mechanical and electronics industry, its abour productivity has been close to the manufacturing sector average. In the case of he electronics industry alone, its labour productivity has persistently outstripped the manufacturing sector by a wide margin since the mid-1980s. These reinforce the refutation of the argument that the industries under consideration might be more labour-intensive than the rest of the manufacturing sector. As a matter of fact, it is customary in trade analysis to treat industries having an above-average value-added 01998 John Wiley Sons, Ltd J.mnt.Dev.10,733-74901998)and Jiang (1993) are among the very few studies in the literature that deal with this topic, both covering mainly the second half of the 1980s. The results which they have obtained do not agree with each other, largely due to their di€erent estimation methodologies as well as di€erent ways of aggregating data in the attempts to achieve correspondence between trade and production categories. A further shortcoming that might seriously limit the usefulness of such studies concerns the comparability of the capital stocks of di€erent industries. It is a salient feature of China's manufacturing sector that some industries (e.g., electronics) are much newer than others. As a result, the value ®gures of capital might obscure the fact that capital productivity di€ers widely across industries, especially because of problems of pricing and depreciation that are inherited from the pre-reform era. In any event, it can be seen from Table 3 that the capital±labour ratio of the mechanical and electronics industry and that of the electronics industry alone have both outstripped the manufacturing sector as a whole, and, in the latter case, the gap has widened over time. Table 4 presents an alternative measure that is simple yet e€ective in indicating the relative capital and technology intensity in production: the per worker net output value of the mechanical and electronics industry, the electronics industry, and the manufacturing sector as a whole. For the mechanical and electronics industry, its labour productivity has been close to the manufacturing sector average. In the case of the electronics industry alone, its labour productivity has persistently outstripped the manufacturing sector by a wide margin since the mid-1980s. These reinforce the refutation of the argument that the industries under consideration might be more labour-intensive than the rest of the manufacturing sector. As a matter of fact, it is customary in trade analysis to treat industries having an above-average value-added Figure 1. Output composition of China's electronics industry, 1983±95 (sectoral shares at 1980 constant-price gross output value). Sources: ZJDGN-D and ZDGN (various years): ZDB (5 March 1996 p. 4). #1998 John Wiley & Sons, Ltd. J. Int. Dev. 10, 733±749 (1998) 738 D. Lo and T. M. H. Chan