Working Paper: February 2000

Safety Nets and Safety Ropes:
Comparing the Dynamic Benefit Incidence of Two Indonesian "JPS" Programs

Sudarno Sumarto (SMERU), Asep Suryahadi (SMERU), Lant Pritchett (World Bank) *

Download Full Report
(272 kbyte, PDF file)

Abstract

Calculations of the benefit incidence and targeting effectiveness of "safety net" programs have typically examined only the relationship between a household's current expenditures and program participation. However, in programs that respond to an economic shock or intend to mitigate household risk, it is not just the current level of expenditures that matters, but also changes in expenditures. While pure "safety net" programs may intend only to benefit those whose are currently poor, programs to mitigate shocks (which we call "safety rope" programs) may intend to provide transfers to those whose incomes have fallen, whether or not they have fallen below an absolute poverty threshold. We examine the targeting performance of two programs created to respond to the social impacts of the crisis in Indonesia. The targeting of each program was different, both in design and in practice. We find strong evidence that one of the programs, a subsidized sale of rice, was targeted to the "permanently" poor while the targeting was only weakly related to the "shock" in consumption expenditures. The employment creation programs were much more responsive to changes in expenditures. A household which began in the third quintile by level of expenditures in 1997 but was in the worst quintile by its fall in expenditures between 1997 and 1998 was four times more likely to have participated in the employment creation program than a household starting from the same level in 1997 but experiencing the most positive shock. In contrast, a household in the middle quintile with the worst shock was only 50 percent more likely to receive subsidized rice.

* We would like to thank BPS and UNICEF for access to the 100 villages data. This paper has benefited from comments and suggestions from Menno Pradhan and Lisa Cameron. We thank Amalia Firman and Yusuf Suharso for their assistance. The remaining errors and weaknesses, however, are solely ours.