The word “dual” has the meaning of double, twofold or in two parts according to The Australian Oxford Dictionary (1996). Applying the word to an economic context, I would describe the term “dual structure of an economy” to indicate the co-existence of two different sectors alongside each other with disparities in technology and productivity between them. Take for instance, the existence of a traditional primary sector along with a modern non-primary sector constitutes to a “dual structure”.
Duality can exist within a particular sector of the economy. The existence of traditional subsistence agriculture alongside export agriculture among the Southeast Asian countries during colonial period is an example of duality within a sector. Minami (1994, p. 239) defined “dual structure” to mean the existence of two distinct sectors; one that has modern technology, a high capital-labour ratio, high labour productivity and wages, against another one, which has all the opposite characteristics.
Therefore, one can infer that if large-scale industries with high capital intensity (K) were promoted at the stage in which labour supply (L) is relatively abundant then differentials in labour productivity (Y/L) and the wage rate would arise and lead to a dual structure. For Japan, three characteristics of the economy indicated the existence of a “dual structure”. First, the coexistence of a very low unemployment rates and continued low incomes in agricultural and small business.
Second, the existence of large (“industrial giants”) and small scale firms within the manufacturing sector. Third, the existence of peculiar characteristics of the labour market such as enterprise unions, wages differentials according to the size of the firm and long term employment (Odaka, 1967 p. 50). Furthermore, duality existed in the manufacturing sector too. In general, “dual structure” of the Japanese economy began before the Second World War as seen in a widening of income differential between agricultural and non-agricultural.
The widening income differential was due to two main reasons. Firstly, productivity differential widen as productivity in the industrial sector increases faster due to relative ease in technology borrowing than productivity in agriculture. Secondly, the existence of surplus labour in traditional industries has tended to hinder increases in productivity due to limited arable land for agriculture so the farmland area per worker increased very slightly.
With reference to Table 1 (Appendix)(Hayami Y, 1997 p. 3), the ratio of real labour productivity in agriculture to that of industry declined from 75% in 1885 to 24% in 1935. However, intersectoral terms of trade remained more or less stable hence the drop is mostly reflected in decreases in the ratio of agricultural income to that of non-agricultural from 76% to 38%. Although there were improvements in the ratio in the post-war period, it could partly be attributed to improvements in terms of trade due to protection policies and an increase in non-agricultural income for agricultural households.
As for the ‘duality’ within the manufacturing sector, it emerged during the interwar period. Several factors namely: age, sex, education and training, work-status and scale of firm could contribute to such wage differentials. In the case of Japan, evidence had suggested that the scale of firm has been very important in Japan (Paine, 1971). As shown in Table 2 and 3 (Appendix), the general trend prior and after the Second World War were the larger the size of capital, higher the wages; the lower the labour productivity, lower the wages and lower the capital productivity, lower the wages.
In all, ‘duality’ developed in response to increased demand for better-educated and trained workers by large-scale industries. The reason being that industrialization reached the stage whereby heavy and chemical industries were the main focus, therefore large-scale industries were willing to pay higher wages to employ better-educated labourers and keep them to internalize investment in the skill formation (Paine, 1971).
To explain for the ‘duality’ of the manufacturing sector in Japan, two theories of dual structure could be of help. The first being Miyazawa-Shinohara’s capital concentration hypothesis, which attributes wage differentials to the differences in average productivity, which are in turn, explained by differentials in capital intensity due to imperfection in capital markets (Odaka, 1967 p. 58). This hypothesis suggests that larger firms were able to obtain bank loans relatively easily as compared to smaller firms.
Moreover, large firms incur a lower cost of capital then smaller firms henceforth with adequate financial capability it was possible for large firms to introduce relatively capital-intensive methods of production thus leading to higher levels of labour productivity. However, this hypothesis required the assumption that large firms draw from a different labour market to small firms to explain persistent wage differentials (Minami, 1994 p. 244).
The second theory is the skill concentration hypothesis developed by Teranishi and Minami, whereby they attempted to explain the ‘duality’ in terms of skilled and unskilled labour. They argued that large firms employ primarily skilled workers while small firms were mainly unskilled workers. In addition, Teranishi and Minami believed that the increase in supply of unskilled labour in the 1920s and the relatively decline in wages to that of skilled labour led to a lower average wages for small firms than large firms.
As a result, wage differentials developed and led to differentials in both the capital-labour ratio and labour productivity because large firms found it profitable to employ labour-saving technology but small firms did not. This second theory is a direct opposite of the Miyazawa-Shinohara’s hypothesis whereby differential in capital-labour ratio leads to differential in labour productivity thus resulted in wage differential. For this skill concentration hypothesis, the assumption that unskilled workers cannot be substituted for skilled workers had to be made (Minami, 1994 p. 4).
The main idea that both theories may be driving at is that a large-scale capital-intensive firm requires better-educated and trained labour to operate new technology machinery and require a stable supply of workers for production operations. But the labour market had only a limited supply of workers with appropriate skills. This led to an increase in the demand for such workers and resulted in the development of the nenko joretsu system.
The nenko joretsu system is whereby large firms select and train certain new entrants to the labour force, offering them guaranteed wage increases as their length of service increased (Paine, 1971 p. 213). In conclusion, there is some truth in both assumptions made in the respective theory. However, both assumptions probably over-simplify the reality. In my own perspective, there exist a combinations of factors that could explain for the ‘dual structure’ within the manufacturing sector although an abundance of labour and the ability to pay of large-scale firms are consistent with the concept of wage differentials.
The abundance of labour does not last forever so as labour surplus decrease, both large and small firms would compete for new workers. Moreover, as technology advances, the introduction of automated production process does not necessary requires skilled workers. Instead the job scope becomes a routine and unskilled workers would be more than able to handle the job (Minami 1994). Despite all the above mentioned, wage differential still prevail in Japan.
In an economy, as long as labourers have different level of skills that requires time to acquire, and the demand for these skills changes with economic progress then some wage differentials would persist (Paine, 1971 p. 234). Furthermore, in Japan, the continued existence of the two unique systems namely; lifetime employment and seniority wage systems. Then differences in wage levels, productivity and capital-labour ratios between large and small firms still exist therefore “dual structure” in the manufacturing in Japan still persist but the differential gap is becoming narrowed.