ABSTRACT
In the 1970s and 1980s, the official unemployment rate was relatively low and stable at 2-3 per cent of the labour force. In the 1990s however, it doubled to 6 per cent, a period otherwise known as the golden age of Indonesia’s economic development. It continued to rise to 10 per cent in the mid-2000s, another period of reasonably healthy economic growth by international standards. The official unemployment rate did abate somewhat in the second half of the 2000s, to 8 per cent in 2008. Though national definitions vary a great deal from country to country, Indonesia’s open unemployment rate in the 2000s compared unfavourably with emerging countries such as Thailand, Malaysia and Mexico (1 to 4 per cent) and was on par with some high income European countries which provided unemployment benefits such as Belgium, Germany and Spain (8 to 10 per cent, The Economist 15-21 March 2008: 117). The three-fold increase in the official unemployment rate between the mid1970s and the mid-2000s, and the relatively high level of underemployment – a third of all employed worked less than 35 hours a week – became serious topics in Indonesia, and often took precedence over income and poverty data. So much so that Central Bureau of Statistics (CBS) conducts a national survey of the labour force situation in all provinces twice a year, in February and August. In addition, it conducts a national socio-economic survey annually in February, which includes unemployment questions. Unemployment questions also feature prominently in the decennial population census, and the mid-decennial population survey. Finally, in the early 2000s CBS began publishing underemployment data in terms of working shorter hours, both voluntarily and involuntarily. Newspaper headlines reported these data as soon as they were released, and opinion pieces regularly warned readers of an impending social crisis. With so much data collected on a regular basis and so much attention from the media, academics and politicians, one would think that the trends and underlying causes of unemployment and underemployment would be pretty much understood. Yet the opposite appeared to be true. This chapter explores the extent to which there was a long-term deterioration in labour market outcomes in the 2000s as portrayed by official unemployment statistics. It finds instead a relative stability or modest movements in a range of other labour market indicators, including employment levels, job search
duration, age and education-specific unemployment rates, the share of youth in total unemployment, and the share of experienced job seekers. The large increases in the official unemployment series were due predominantly to significant changes in definition in the mid-1990s and again in the early 2000s. Using a consistent definition, open unemployment can be shown to have remained stable during most of the 1990s until the onset of the 1997-1999 financial crisis. Thereafter, unemployment rose gradually in the 2000s, mainly as a result of the rising education level of rural job seekers. These, as in the case of their urban counterparts, increasingly aspired to more formal sector jobs requiring longer job search durations. The chapter also reviews the trends and factors responsible for educated and youth unemployment, and gender and provincial disparities. A major theme running throughout the chapter is the need to exercise a great deal of caution in the use of published unemployment statistics, to use unemployment series using a consistent definition, and to combine this with other labour market indicators, before committing to policy action. In particular, the assessment of unemployment levels and trends must consider the continuous flow of new entrants in the ranks of job seekers in a growing labour force throughout the year, and the eventual exit of most of them from unemployment figures after a period of time, as well as the stock of the long-term unemployed at any point in time. The chapter argues that the major unemployment policy issue may not be a mismatch between job seekers and the requirements of employers, though this may be the case to a certain extent, but transitional unemployment dominated by first-time job seekers. These spent around eight months on average to find work, a considerable period of time before they eventually succeeded. Appropriate measures to reduce job search duration would result in less hardship, anxiety and financial burden in their transition from education to work. Some of the measures are suggested under the rubric of proactive policies in the final Chapter 7.
