Varying estimates of poverty have often resulted in some deprived communities being excluded from government welfare schemes. Sabarwal and Chowdhry look at the case of Satat Jeevikoparjan Yojana in Bihar, which uses the mechanisms recommended by BRAC’s Graduation Approach and harnesses community knowledge to identify those living in extreme poverty and ensure that they receive access to social protection schemes. They discuss how this targeting approach takes into account non-monetary deprivation, and ensures the inclusion of women and marginalised groups.

In India, the definition and estimates of poverty have been long debated. The historical lack of consensus around measuring and defining poverty has led the government and non-government organisations to use varying methodologies to identify and design solutions for millions of people facing extreme poverty.

In recent times, these varying approaches have led to starkly different measures of poverty: in April 2022, an International Monetary Fund (IMF) report authored by Surjit Bhalla, Karan Bhasin and Arvind Virmani estimated that only 0.8% of India’s population lived in extreme poverty – indicative of near eradication – while World Bank estimates found this figure to be at 10%. By this estimate, approximately 130 million people live in extreme poverty, with an additional 56 million estimated to have slipped into poverty after the onset of the pandemic in 2020. Simultaneously, NITI Aayog’s Multidimensional Poverty Index estimates, launched in 2021, classified 25.01% percent of India’s population as living in poverty along multiple parameters.

The effectiveness and accuracy of any approach, however, is dependent on the collection of regular and rigorous data. This is difficult to implement on a large scale, often resulting in the persistence of inclusion and exclusion errors – the former referring to the provisioning of social support for the non-needy, and the latter to inadequate coverage or failure to reach those who are eligible for these schemes.

Satat Jeevikoparjan Yojana (SJY), an ongoing scheme launched by the Government of Bihar, is using a novel tool to address this challenge and identify those living in extreme poverty: community knowledge. Tapping into the identification processes adopted by Bihar’s extensive community-based organisations has helped limit exclusion errors and to extend livelihood support to women across 200,000 ultra-poor¹ rural households in the state.

Can gaps in India’s poverty data thus be bridged by implementing alternative approaches like leveraging community knowledge of local conditions and contextual nuances at scale? Experiences from Bihar suggest that it might.

Establishing a more nuanced definition of poverty

Typically, poverty has been measured based on a monetary threshold below with which a person’s needs cannot be met. Most prominently, the World Bank’s threshold of $2.15 per person per day has been used to define national levels of extreme poverty in some of the world’s poorest countries.

While there are multiple definitions of poverty, globally, it is agreed that poverty is a state of multiple and simultaneous deprivation of imminent needs². This encourages policymakers and researchers to not only move beyond the monolith of monetary poverty, but also adopt a more nuanced approach to understand the relative and heterogeneous nature of the non-monetary deprivations that determine the overall wellbeing of households. In India, efforts to move beyond monetary measures alone are exemplified by the Socio Economic and Caste Census (last conducted in 2011) and the Multidimensional Poverty Index (based on the National Family Health Survey-4, conducted in 2015-16), which aim to capture multiple and overlapping deprivations.

Non-monetary deprivations are often rooted in specific socioeconomic, political, and cultural contexts, in addition to other factors such as existing local infrastructure. Often, these conditions vary even within a single state – for example, the characteristics of a household experiencing extreme poverty in coastal Tamil Nadu will be different from those of one located in the hilly region of the state. Commonly used measures of poverty, such as access to household appliances like fans, might result in incorrectly excluding households in warm coastal regions (which are more likely to have fans due to greater necessity) from poverty estimates.

The Graduation Approach, a sustainable livelihoods programme first developed by BRAC, an NGO in Bangladesh, offers a potential solution to minimise inaccuracies while identifying ultra-poor households. The programme was first implemented in 2006 to address the needs of those living in extreme poverty, without access to regular microfinance services and government welfare programmes.

Made up of six complementary and sequential components (including a productive asset, training, coaching, access to savings, and consumption support), each designed to address specific deprivations faced by ultra-poor households, the programme is implemented over two years and aims to provide a ‘big push’ to help those living in ultra-poverty transition to more secure livelihoods. To date, more than 100 organisations have adapted the approach in 50 different countries to reach 14 million people. 

Targeting ultra-poor households under the Graduation Approach 

One of the reasons behind the success of this programme is the targeting method employed by the approach, which involves a rigorous process that captures eligible households and minimises exclusion. These targeting mechanisms or eligibility criteria include a combination of national poverty data, existing data on households, community knowledge, and surveys administered by programme staff. This is further supplemented by setting up a verification process that ensures that all criteria are met. 

Through a high level of community involvement and engagement with community influencers such as village leaders and village members, the Graduation Approach employs effective targeting that mitigates barriers to gender equity and ensures the inclusion of marginalised groups by taking into account their earnings potential (Alatas et al. 2012). 

Results from an evaluation of the Graduation Approach across six countries show that the programme has been successful in identifying and targeting ultra-poor households – the beneficiary group mainly comprised of households living below US$1.25 per day (or below the poverty line) (Abdul Latif Jameel Poverty Action Lab (J-PAL), 2015). 

However, is it possible to replicate this process at scale? 

Identifying the extreme poor in Bihar, at scale 

In 2019, the Government of Bihar launched the Satat Jeevikoparjan Yojana, a social protection scheme to help the poorest of the poor break out of extreme poverty. Under SJY, the Bihar State Rural Livelihood Mission (JEEViKA) is currently using and adapting the mechanisms recommended by the Graduation Approach to target 200,000 ultra-poor households. 

For rural households to be considered ‘ultra-poor’ under SJY, the programme considers twelve distinct indicators (other than income, caste, and occupation) to understand their social and economic conditions. These include indicators of the presence of any kind of savings, access to government schemes that the household is eligible for, active participation in social groups (Self-Help Groups (SHGs) or other community-based organisations), access to formal loans, and the educational status of school-going children in the households, among other factors. 

Based on the targeting approach used in a pilot study conducted in collaboration with Bandhan-Konnagar³, SJY makes use of transect walks⁴, social mapping, and wealth ranking to identify ultra-poor households based on these criteria. Crucially, it taps into prevailing community knowledge and leverages inputs from existing community-based organisations and cadres (such as SHGs, Community Resource Persons (CRPs), and other village organisations) to identify and rank ultra-poor households living in the most extreme conditions of poverty. Block-level JEEViKA officials then carry out a final check of the endorsed list of SJY participants to avoid instances of elite capture (Samaranayake et al. 2021). Special identification drives are then conducted as follow-ups to ensure that no one is left behind. 

This approach towards identifying and targeting individuals living in extreme poverty, thus, promotes community ownership over the process and also ensures timely updates to these figures – a feat enabled by tapping into community knowledge of local conditions. Process monitoring data also show minimal exclusion errors. 

SJY’s alternative approach is also helping correct exclusion errors that have typically prevented women from ultra-poor households from accessing community organisations such as SHGs. In India, among the 8 million registered SHGs, the highest proportion (55%) of members are reported to belong to communities other than Scheduled Castes, Scheduled Tribes or other minority communities. This is despite the fact that, in India, 5 out of 6 multidimensionally poor are from lower tribes or castes. 

Linking households to existing social protection schemes 

The effectiveness of any social welfare scheme rests on its ability to reach its intended beneficiaries. Often, undercounting through exclusion errors results in restricted access to the myriad schemes that have been launched to support those who need them most, keeping them trapped in poverty. For example,  a report by the National Sample Survey Organisation (NSSO) found that, among the rural and urban households with the lowest monthly per capita expenditure (MPCE), only 41% and 29% held Below Poverty Line (BPL) ration cards. 

Lessons from the ongoing implementation of SJY in Bihar suggest that the Graduation Approach not only helps accurately identify ultra-poor households by buttressing monetary measures with contextual knowledge, but also helps link these households to mainstream other social protection schemes in the process (Markhof 2020). 

Regular process monitoring surveys were conducted over five years (2019-2023), with over 11,000 participants, to study the programme’s evolution. Data from surveys conducted in 2020 highlight accurate targeting of households in accordance with the scheme’s eligibility criteria – 6% of the surveyed beneficiary households were involved in toddy production⁵, 47.43% belonged to SC/ST communities, and 52.26% belonged to other backward classes. Additionally, 42.31% of surveyed households reported having no income in the past year. Of the remaining 56.41%, the average annual income was reported to be Rs. 9,679.55 (or approximately Rs. 807 per month).   

Community-based targeting goes beyond simply identifying those who need support, but by providing the ultra-poor access to a range of not previously available to them, it also serves as a powerful pathway of making the invisible, visible. In Bihar, for example, under SJY, bank accounts have been opened for participants; participants are linked to the national Public Distribution System (PDS) scheme; and eligible households have been enrolled under Pradhan Mantri Suraksha Bima Yojana (PMSBY) and Pradhan Mantri Jeevan Jyoti Bima Yojana (PMJJBY).⁶

The way forward

Working directly with individuals and households through community-based organisations can be an effective pathway for identifying and targeting beneficiaries at scale. The case of Bihar shows that support in the form of a model like the Graduation Approach can help the ultra-poor move out of poverty, and to more sustainable livelihoods.

Notes:

1. Ultra-poor households live in conditions of extreme poverty on less than half of the global poverty line (BRAC, 2013).
2. The global Multidimensional Poverty Index (MPI) is an international measure of acute poverty covering over 100 developing countries. It complements traditional income-based poverty measures by capturing the deprivations that each person faces at the same time with respect to education, health and living standards.
3. Bandhan-Konnagar is a non-profit organisation which launched Targeting the Hard-Core Poor (THP) based on the Graduation Approach in 2007. They have since expanded to reach over 160,000 households across India, and also provide technical assistance on the Bihar scale-up of the programme.
4. Transect walks are defined by the World Bank as a tool for describing and showing the location and distribution of resources, features, landscapes, and main land uses along a given transect, or fixed line along a landscape.
5. The alcohol ban in Bihar in 2016 had a disproportionate impact on traditional toddy producing communities. One of the eligibility criteria at the time of launch of this programme was to include ultra-poor households from toddy tapping families.
6. PMSBY is an accident insurance scheme offering death and disability cover to individuals between 18 and 70 years; PMJJBY similarly offers life insurance cover for death due to any reason for individuals between 18 and 50 years.

Further Reading

• Alatas, Vivi, Abhijit Banerjee, Rema Hanna, Benjamin A Olken and Julia Tobias (2012), “Targeting the Poor: Evidence from a Field Experiment in Indonesia,” American Economic Review, 102(4): 1206-1240. Available here.
• Markhof, Y (2020), ‘Economic Inclusion and the National Rural Livelihoods Mission (NRLM) Jeevika’, socialprotection.org, 3 June.
• Samaranayake, S, P Singh, A Ranjan, K Guha and G Patel (2022), ‘Case Study 2: The State of Bihar’s Approach to Economic Inclusion: JEEViKA and the SJY Program’, World Bank. Available here.
• Abdul Latif Jameel Poverty Action Lab (2015), ‘Building stable livelihoods for the ultra-poor’, J-PAL Policy Insights.
• BRAC (2013), ‘Ending extreme poverty’, BRAC Briefing Note #1. .