Designing Anti-Poverty Programs in Emerging Economies in the 21st Century: Lessons from Indonesia for the World

Abstract

Governments of developing countries around the world have dramatically expanded social protection programs for the poor in recent decades. In doing so, they face a host of challenges in the targeting, design and implementation of these programs. In this paper, I describe the results from more than a decade of collaboration with the Indonesian government to understand how best to tackle these challenges, drawing primarily on evidence from randomised controlled trials. I highlight results that show the advantages of both community-based targeting and self-targeting, the importance of tangible information about beneficiaries’ rights in minimising leakage, and the remarkable impacts of conditional cash transfers in the medium term. I also describe several recent studies that use randomisation at scale to generate policy-relevant evidence.

Pemerintah negara berkembang di dunia secara dramatis telah mengembangkan program perlindungan sosial untuk kaum miskin dalam dekade terakhir. Dalam menyusun program, tantangan yang dihadapi adalah menentukan sasaran, desain dan implementasi. Dalam tulisan ini, saya menguraikan hasil dari sepuluh tahun lebih kerjasama dengan pemerintah Indonesia untuk memahami cara terbaik mengatasi masalah-masalah ini, dengan mengambil bukti terutama dari metode uji acak terkendali (randomized controlled trials). Saya menggarisbawahi temuan akan pentingnya sasaran berbasis komunitas dan berbasis pribadi, pentingnya informasi nyata mengenai hak penerima manfaat dalam meminimalkan kebocoran, dan dampak luar biasa dari bantuan tunai bersyarat, yang diberikan dalam jangka menengah. Saya juga menguraikan beberapa studi terkini yang mengunakan randomisasi dengan skala yang tepat untuk menghasilkan bukti yang relevan untuk kebijakan.

ACKNOWLEDGEMENTS

The results described in this paper draw on my collaborations with many people, including Vivi Alatas, Abhijit Banerjee, Nur Cahyadi, Gabriel Kreindler, Jordan Kyle, Junko Onishi, Rizal Adi Prima, Ririn Purnamasari, Audrey Sacks, Elan Satriawan, Sudarno Sumarto, Ekki Syamsulhakim, Julia Tobias, Matthew Wai-Poi, Susan Wong and particularly Rema Hanna. I thank our many partners in the Indonesian government, particularly Rudy Prawiradinata, Pungky Sumadi, Maliki and Vivi Yulaswati at Bappenas, Bambang Widianto at TNP2K, and many partners at BPS, the Ministry of Home Affairs and the Ministry of Social Affairs—for their openness to using evidence to inform policy decisions and for their support of the projects described here. I thank the many staff members at J-PAL Southeast Asia for outstanding work implementing these projects, including Talitha Chairunissa, Masyhur Hilmy, Amri Ilmma, Nurzanty Khadijah, Chaerudin Kodir, Nadira Purdayinta, Héctor Salazar Salame, Ritwiki Sarkar, Prani Sastiono, Freida Siregar, Jurist Tan, Poppy Widyasari and particularly Lina Marliani; Gregorius Kelik Agus Endarso and Yulia Herawati for their help with the Generasi project; and SurveyMeter and Mitra Samya for their partnership on many of these projects. I also thank Yuen Ho, Evelina Smirnitskaya and Claire Walsh for help preparing this paper. Finally, I thank Sudarno Sumarto for hosting me as a young graduate student at SMERU, studying the OPK/Raskin program in the summer of 2001; Lant Pritchett for introducing me to some of the challenges of social assistance in Indonesia in the fall of 2000; the Henry Luce Foundation for sending me to Indonesia in the first place in 1997–98; and Scott Guggenheim for almost two decades of unwavering support and encouragement.

Funding for the projects described here came from the Australian government (AusAID and DFAT), the Royal Netherlands Embassy, 3ie and the World Bank. All the views expressed in the paper are mine and do not necessarily reflect the views of any of the institutions or individuals acknowledged here.

Notes

1 Rastra (Rice for the Prosperous Population program); BPNT (e-voucher, non-cash component of Rastra); Raskin (Rice for the Poor program); OPK (Special Market Operation program); BLT (Direct Cash Assistance program); BLSM (Temporary Direct Community Assistance program); PKH (Hopeful Families Program); BSM (Poor Students Assistance program). See glossary for the Indonesian names of these programs.

2 See Aker (2017); Banerjee et al. (2015); Banerjee et al. (2017); Blattman, Jamison and Sheridan (2017); Evans and Popova (2016); and Haushofer and Shapiro (2016).