Targeted information to improve social assistance
Identification cards printed with information about program eligibility and entitlements present a potentially simple, low-cost way to improve access to targeted social assistance programs. Research by J-PAL affiliates Abhijit Banerjee (MIT), Rema Hanna (Harvard), and Benjamin Olken (MIT), together with Jordan Kyle and Sudarno Sumarto, has shown that providing ID cards to beneficiary households, in combination with community outreach and advertisements, improved access to Indonesia’s national rice subsidy program (known as Raskin). Drawing on this research, the Government of Indonesia decided to scale up social assistance ID cards to the poorest households across the country. In June 2013, the government began distributing ID cards to 15.5 million of the poorest households, reaching 65.67 million people.
Corruption prevents social assistance programs from reaching intended beneficiaries.
Raskin is Indonesia’s largest permanent, targeted social assistance program. In 2012, the budget for Raskin was US$ 1.5 billion and the program provided subsidized rice to households in the bottom 30 percent of the income distribution. Households eligible for Raskin can purchase 15 kilograms of rice per month at a subsidized price of one-fifth the market value. However, in practice, eligible households received only one-third of the amount of rice they were entitled to at a cost that was 42 percent more than the official subsidy price. Leakages are also high: many households that purchase subsidized rice are ineligible for the program. Additionally, beneficiaries were not always aware of program rules and eligibility requirements. While local officials were supposed to use an official roster of eligible households to determine who could access the program, they sometimes used their own discretion instead.
In 2012, the Government of Indonesia began to explore reforms to Raskin for the first time since the program’s inception in 1998. The National Team for Accelerating Poverty Reduction (known as TNP2K), under the direction of the Vice President of Indonesia, proposed an ID card system for the targeted social assistance program.
ID cards could empower poor households to demand the full subsidy to which they are entitled under the program. They could also give distributors an easy way to verify if households are eligible. Conversely, a rigid ID card system could lead to conflict within communities if the eligibility process is perceived to be unfair or inequitable. As such, the Vice President voiced concerns and requested that the team put together real evidence about the effectiveness and social implications of distributing cards.
Identification cards coupled with enhanced community engagement increased the share of eligible households purchasing Raskin rice at the correct price.
J-PAL affiliates and colleagues worked closely with TNP2K to design a randomized evaluation that could generate evidence on the proposed identification cards system in time for the government to incorporate the program into the national budget for 2013 if found effective. Specifically, they measured how the information printed on the cards and the channels through which the cards were distributed affected program outcomes. Trained program facilitators also conducted enhanced community engagement in randomly selected villages to address the potential for social conflict within communities. They met with village leaders, hung informative posters throughout the village, and read scripted messages about the program on local village announcement systems.
Within six months, researchers collected and analyzed data and found that the ID cards increased monthly purchases of Raskin rice by 1.25 kg (or 24 percent), and decreased price mark-ups by IDR 57/kg (2.5 percent), therefore increasing the monthly subsidy eligible beneficiaries received by IDR 7,455 (26 percent). The community socialization component increased beneficiary take-up by an additional 0.81 kg of rice per eligible household and improved overall beneficiary satisfaction with the Raskin program.
These impacts were achieved at a cost of US$1 per household for the ID cards, and US$1.40 per household for the community engagement. The ID card and community engagement interventions led to an effective subsidy increase per household per year of US$7.71 and US$4.58, respectively.
We need to and are committed to showing that our policies actually benefit the poor…The critical ingredient was a partnership with researchers from the start.
–Sudarno Sumarto, policy adviser, TNP2K, Office of the Vice President of the Republic of Indonesia
For more details, see the evaluation summary.
From Research to Action
Rapid research in response to policymaker demand created buy-in for evidence from the beginning.
The randomized evaluation not only showed that the ID cards were effective, but also that the cards did not generate social conflict between beneficiaries and non-beneficiaries in the recipient villages. According to Bambang Widianto, (executive secretary of TNP2K), this study provided key insights into the design of Raskin cards and its socialization strategy.
The Government of Indonesia used insights from this research to directly scale up a program in the same context in which it was evaluated. In June 2013, the government decided to scale up the ID card program to 15.5 million poor households. By August of that year, 14.38 million cards had been distributed as Social Protection Cards.
Watch Bambang Widianto and Rema Hanna discuss the evaluation and scale up:
Although the cards in the evaluation were intended for the Raskin program, the government leveraged the research findings to integrate access to two additional programs into the Social Protection Card program: Bantuan Langsung Sementara Masyarakat (BLSM), a temporary unconditional cash transfer program designed to compensate for cuts in Indonesia’s fuel subsidy, and Bantuan Siswa Miskin (BSM), a cash transfer for poor students.
In total, these three programs cost approximately US$4.3 billion. Importantly, as part of the scale-up, the government implemented a community engagement process to allow households to reallocate cards for individuals who moved or whom the community deemed as too rich to receive the program; this community process was designed based on a previous evaluation that J-PAL affiliates had conducted with the Government of Indonesia.
In 2016, the Indonesian government announced a reform to deliver Raskin (now called Rastra) and five other major social protection programs as electronic voucher programs by 2022. The government will distribute Rastra benefits as electronic vouchers deposited directly into eligible households' accounts on the 25th of each month. These vouchers will be accepted as a means to purchase rice and eggs at registered private retailers. J-PAL affiliated researchers and coauthors are now working with the government to evaluate the overall impact of this reform and variations in its implementation.
Abhijit Banerjee, Rema Hanna, Jordan Kyle, Benjamin A. Olken, and Sudarno Sumarto, 2018. "Tangible Information and Citizen Empowerment: Identification Cards and Food Subsidy Programs in Indonesia." Journal of Political Economy 126 (2): 451-491.https://doi.org/10.1086/696226.
Banerjee, Abhijit, Rema Hanna, Jordan Kyle, Benjamin A. Olken, and Sudarno Sumarto. "The Role of Competition in Effective Outsourcing: Subsidized Food Distribution in Indonesia." Working Paper, March 2017.https://www.povertyactionlab.org/publication/role-competition-effective-outsourcing-subsidized-food-distribution-indonesia
Abdul Latif Jameel Poverty Action Lab (J-PAL). 2018. "Targeted information to improve social assistance." J-PAL Evidence to Policy Case Study. Last modified November 2019.