INTRODUCTION

In the middle of 1997 Indonesia was hit by a major economic crisis. The social impact of the large economic contraction that occurred was very substantial, reversing many of the improvements in the social sector that had been achieved in previous decades. In the labour market, although the open unemployment rate only slightly increased from 4.7 per cent in 1997 to 5.5 per cent in 1998, real wages fell by approximately one-third (Feridhanusetyawan 1999; Manning 2000). The national headcount rate of poverty, according to one estimate, increased from 15.7 per cent in 1996 to 27.1 per cent in 1999 (Pradhan et al. 2001). Another study, which tracked the poverty rate over the course of the crisis, estimated that the poverty rate increased by 164 per cent from the onset of the crisis in mid-1997 to the peak of the crisis at the end of 1998 (Suryahadi et al. 2000).

It is obvious that poverty increased rapidly along with the worsening of the crisis, implying that there were large numbers of households moving into poverty in a short period of time (Suryahadi et al. 2000; Skoufias et al. 2000). In contrast to the static concept of the headcount rate of poverty, the fact that the poverty rate changed relatively quickly over short periods of time following the crisis shows that, in reality, the state of poverty is dynamic.

Although the dynamic nature of poverty only recently gained wide attention, it is not a new concept. It was already well known that, over time, households enter into and exit from poverty (Bane and Ellwood 1983). This means that there is always a chance that people who are currently not poor will fall below the poverty line and at the same time that the poor will move out of poverty at some time in the future. This leads to the concept of “vulnerability to poverty”, defined as the risk that a household will be poor in the near future.

The economic crisis brought severe multiple shocks to Indonesian households. The present study is an attempt to assess what happened to poverty and vulnerability to poverty in Indonesia before and after the economic crisis. The analysis is based on a method specifically developed for estimating vulnerability to poverty using cross-sectional data.