The government of Indonesia incorporated community-based targeting (CBT) in its emergency cash transfer programmes in response to the 2007 financial crisis and more recently, in response to COVID-19. CBT involves consulting community members to identify the poorest households eligible to receive cash transfers. Under the right conditions, community-based targeting can be used to quickly and effectively reach households that might otherwise be excluded through more conventional targeting approaches.
In 2005, the government introduced a temporary unconditional cash transfer under the BLT cash assistance programme (Bantuan Langsung Tunai) to help households cope with fuel subsidy cuts. A proportion of government savings from removing the subsidy was channeled into the new transfer programme that reached 18.5 million households.1 Recipient households were identified using proxy means testing (PMT) drawing on newly collected census data.2 At the time, the transfer programme was highly controversial, and widespread dissatisfaction with the initial beneficiary list contributed to protests and the resignation of some village leaders.
In 2008, the government introduced additional transfers in response to the global economic crisis. To address concerns about exclusionary targeting of the 2005 programme, community-level verification of beneficiary lists was introduced, a process known as community-based targeting (CBT). Within the programme, a field experiment was conducted to test the performance and popular acceptance of different targeting approaches. The study examined PMT, CBT and a hybrid approach using CBT to limit the eligibility list and PMT to check for elite capture. Results showed that CBT led to increased acceptance and perceived legitimacy of social protection distribution and 60 percent fewer complaints than PMT.3
In response to the COVID-19 pandemic, the government expanded social protection to an additional nine million households not previously receiving support through major national programmes to expand support through the pandemic. These transfers needed to be disbursed quickly to be effective. Building on earlier targeting experiences, community volunteers were appointed to help identify and reach cash transfer beneficiaries.
Implementation
In 2005, the government of Indonesia developed a registry of poor and vulnerable households through a survey of 19 million households in the bottom three deciles of the income distribution. The results of this survey were used to develop the country’s first unified targeting approach for social protection, first known as PSE05 (Pendataan Sosial Ekonomi Penduduk 2005) and later revised to become the Unified Targeting System (UDB).
The 2008 restructuring of the BLT program updated this list using community verification. The programme targeted poor and near-poor households with three transfers over nine months equivalent to around 15 percent of the average monthly household consumption budget. In villages where CBT was used, community meetings were facilitated to determine local characteristics of poverty and households in the village were ranked by participants in these meetings according to these characteristics. Households below a set quota were deemed eligible for the transfer programme. The hybrid method started with the same procedure, followed by verification against household PMT scores based on the existing beneficiary registry. The 2008 transfers reached a similar proportion of the poorest households as the 2005 transfers, though evidence from the field experiment found that targeting using CBT and a hybrid CBT and PMT approach led to higher public approval of the transfer program.
Among newly introduced programmes intended to reach new beneficiaries impacted by the COVID-19 pandemic, the BLT-Dana Desa (BLT-DD) Village Fund Cash Transfer involved a CBT component. The unconditional cash transfer of IDR 600,000 per month (around USD 40) for three months was intended to reach eight million households that were not already receiving social protection assistance prior to the pandemic. Beneficiary lists were finalized through village discussions and volunteers were appointed at the neighborhood level to assist in the distribution of cash transfers.4
Cost
The original 2005 BLT cash transfer program was tax funded, primarily through diverted funds from the fuel subsidy program. The total value of the program in 2008-09 was USD 1.8 billion, or approximately USD 5 per recipient.5 The BLT-DD cash transfer program made up an estimated 31.2 percent of the total COVID-19 social assistance response budget of IDR 205.6 trillion, or around USD 14 billion.6
Assessment
Inclusive targeting of social protection is challenging in all contexts, not least in a country as large and diverse as Indonesia. Regularly updated information is needed on household circumstances and proxy categories establishing qualifying households can overlook local factors that contribute to poverty and vulnerability, such as social exclusion of particular groups or living costs determined by local factors.
Evidence from Indonesia’s experimentation with community-based targeting shows this approach has some limitations, including the potential for elite capture and exclusion of eligible beneficiaries. However key benefits have been identified, including popular support for social protection programming and inclusive and effective targeting if applied well. The latest application of CBT as part of Indonesia’s social protection expansion in response to COVID-19 had challenges, but is broadly seen as a success. In 2020, an estimated 86.7 percent of households received some form of social assistance, and the BLT-DD programme that used CBT was found to be among the more progressive of Indonesia’s cash transfer programmes in response to the pandemic.7
Drawing on local knowledge to inform social protection targeting through CBT has the potential to contribute to progressive targeting outcomes, legitimacy and community satisfaction.8
References
- 1. World Bank, Jakarta Office, “BLT Temporary Unconditional Cash Transfer: Social Assistance Program and Public Expenditure Review 2,” World Bank, https://openknowledge.worldbank.org/bitstream/handle/10986/26698/673240WP0BLT0T00Box367866B00PUBLIC0.pdf?sequence=1&isAllowed=y.
- 2. Proxy means testing is used where information about household income cannot be easily obtained to determine social protection eligibility. Data from household surveys are used to predict household welfare based on a set of proxies such as age, number of household members, level of education and housing characteristics.
- 3. Alatas, V. et al., “Targeting the Poor: Evidence from a Field Experiment in Indonesia,” The American Economic Review 10, no. 24 (2012): p.1206-1240, https://www.aeaweb.org/articles?id=10.1257/aer.102.4.1206.
- 4. Abdul Latif Jameel Poverty Action Lab (J-PAL), “Community-based targeting to combat Covid-19-induced poverty,” 2021, https://www.povertyactionlab.org/case-study/community-based-targeting-combat-covid-19-induced-poverty.
- 5. World Bank, Jakarta Office, “BLT Temporary.”
- 6. Hebbar, M., Satriana, S. & Ammon, J., “Using social protection to respond to the COVID-19 pandemic in Indonesia,” DAI Global UK Ltd, United Kingdom, 2021.
- 7. UNICEF, UNDP, Prospera, and SMERU, “The Social and Economic Impact of COVID-19 on Households in Indonesia: A Second Round of Surveys in 2022, Jakarta, Indonesia,” 2022, https://smeru.or.id/en/publication/social-and-economic-impact-covid-19-households-indonesia-second-round-surveys-2022.
- 8. McCord, A., “Community-based targeting in the Social Protection sector,” ODI Working paper 514, 2013, http://cdn-odi-production.s3-website-eu-west-1.amazonaws.com/media/documents/11666.pdf.