Quantifying Vulnerability to Poverty: A Proposed Measure, Applied to Indonesia

Vulnerability is an important aspect of households’ experience of poverty. Many households, while not currently “in poverty” recognize that they are vulnerable to events that could easily push them into poverty — a bad harvest, a lost job, an unexpected expense, an illness, an economic downturn. Most operational measures define poverty as some function of the shortfall of current income or consumption expenditures from a poverty line, and hence measure only poverty at a single point in time. We propose a simple expansion of these measures to quantify “vulnerability” to poverty. We define vulnerability as a probability, the risk a household will experience at least one episode of poverty in the near future. A household is defined to be vulnerable if it has 50-50 odds or worse of falling into poverty. Using these definitions we calculate the “Vulnerability to Poverty Line” (VPL) as the level of expenditures below which a household is vulnerable to poverty. This VPL allows the calculation of “Headcount Vulnerable Rate,” the proportion of households vulnerable to poverty, which is the direct analogue of the “Headcount Poverty Rate.” We implement this approach using two panel data sets from Indonesia. We first show that if poverty line is set so that the headcount poverty rate is 20 percent, the proportion of households that are vulnerable to poverty is around 30 to 50 percent. So in addition to the 20 percent that are currently poor (hence are by definitions vulnerable to poverty), an additional 10 to 30 percent of the population is at substantial risk of poverty. Second, we illustrate the usefulness of this approach by examining differences in vulnerability between households by gender, level of education, urban-rural areas, land holding status, and sector of occupation of the household head. The conclusion speculates on the policy implications of these high levels of vulnerability.