The Indonesian manufacturing sector does not seem to display the normal process of positive de-industrialisation as the economy progresses, as in the case of advanced economies. In the past decade, four key changes have characterised depressed development within the sector. First, there has been a significant decline in the manufacturing sector’s contribution to regular waged employment with a slight increase in its contribution to overall employment. Second, there has been an increased level of casualisation of employment. Third, real wages of regular employees have been depressed, in relative term, compared to those in other sectors. Fourth, the manufacturing sector’s share in total exports has fallen, reversing the trend observed during the Suharto era when manufacturing exports increased dramatically.

As a result, the case for re-industrialisation has been widely advocated, including by institutions such as the World Bank.

A widening gap between earnings and labour productivity has also been observed within the manufacturing sector. The decoupling of earnings and productivity is evident economy-wide in post-crisis and decentralised Indonesia — across sectors and provinces. The situation in Indonesia mimics the International Labour Organization’s global finding on this phenomenon, which has been taking place for the past decade.

This delinking is at odds with standard economic theory which posits that real earnings and labour productivity should move in a similar direction. Higher labour productivity should lead to higher real earning and vice versa.

The gap between earning and productivity in Indonesia’s manufacturing sector is quite striking. In 2012, earnings in manufacturing were only 12 per cent of labour productivity, a decline of 2.4 percentage points, from 14.4 per cent in 2001. Among nine economic sectors, the manufacturing sector’s earnings to productivity ratio was the lowest and showed a declining trend. This suggests that the delinking trend between earning and productivity in the manufacturing sector is quite significant.

A positive association between real earning and employment indicates productivity-driven employment growth. That is, in a dynamic economy, an improvement in productivity translates into an increase in labour demand and higher productivity enables increased employment at a higher wage rate. Thus, one may expect that employment, real earning, productivity and the overall GDP, would all move in the same direction. This refers to an ideal situation, where real earning and the quality of employment increase, while overall employment and the economy expand.

But when we analyse the data, we see that there is a negative correlation between growth in employment and growth in wages. One explanation for this phenomenon may be due to the size of firms: in general, workers at large and medium firms tend to be paid a larger share of the fruits of their labour than do workers at smaller firms.

But in large- and medium-scale manufacturing, increases in employment tend to go hand in hand with higher wages. This is the situation that one would expect from microeconomic theory. However, the employment share of large and medium firms in the overall manufacturing sector has declined from 36 per cent in 2001 to 32 per cent in 2011.

A key implication of this concerns the importance of re-industrialisation for quality employment, which is vital for the overall upgrading of the economy, especially the manufacturing sector. The World Bank recently stressed the importance of Indonesia’s manufacturing sector for its economy, but the bank’s policy prescription has not changed much from the its well-known policy view with regard to industrialisation in developing countries. The World Bank’s East Asian Miracle report is the best example of an argument advocating that ‘difficult’ policies like (selective) industrial policy should not be tried by developing countries with limited bureaucratic capabilities.

However, as the experience of earlier industrialised countries in East Asia shows, the role of the government in industrial development cannot be overlooked. As Ha-Joon Chang has argued, the government can pick ‘winners’ and use ‘selective’ policy interventions in the form of policy packages to create and support ‘winners’. Thus, there should be momentum for renewed industrial policy to reverse the trend of de-industrialisation that has troubled Indonesia since the Asian crisis of the 1990s.

Mohammad Zulfan Tadjoeddin is a Senior lecturer in development studies at the University of Western Sydney.