evidoria

← Back to browse

Good practice Imported

Aswesuma — Sri Lanka's Algorithmic Welfare-Targeting Score, and Its Exclusion Problem

Sri Lanka · Colombo · See the Sri Lanka profile · See the Colombo profile

Evidence: Observational / pre–post Top 83% 40/100 · Ask Evidence Copilot about this practice

Sri Lanka's Aswesuma programme uses a 22-indicator Multi-dimensional Deprivation Score computed by the Department of Census and Statistics to sort welfare applicants into benefit tiers; independent reviews document high exclusion of eligible households and low transparency in how

18.8 %
Population coverage under Samurdhi (2016)
29.1 %
Population coverage under Aswesuma (2024-25)
35 %
Government coverage target
12.34 million (55.7%) of population
Population assessed as multi-dimensionally vulnerable
~2 million households
Households selected as beneficiaries

Details

Maturity
Established
Promoter
Welfare Benefits Board / Department of Census and Statistics, Ministry of Finance, Planning and Economic Development
Period
2023-2025
Keywords
social protection, welfare targeting, algorithmic scoring, poverty measurement

Context

Aswesuma is Sri Lanka's formula-based welfare-targeting programme, launched in 2023 as the IMF-linked successor to the discretionary Samurdhi scheme. Sri Lanka's Department of Census and Statistics computes a Multi-dimensional Deprivation Score from 22 indicators gathered via household survey, sorting households into one of four benefit tiers.

Objectives

Replace manual and politically discretionary welfare targeting with an objective, formula-based scoring system.

Activities

Enumerators gather household data on education, health, housing, assets, demographics and economic status using Proxy Means Test methodology to compute each household's deprivation score.

Results

Coverage rose from 18.8% of the population under Samurdhi in 2016 to 29.1% under Aswesuma per the 2024-25 BRIGHT survey, still short of the government's 35% target; separately, Groundviews found 12.34 million people (55.7% of the population) were assessed as multi-dimensionally vulnerable, yet only around 2 million households were actually selected as beneficiaries.

Conclusions

Human Rights Watch documented 'chaotic' reform outcomes with arbitrary cutoffs and high exclusion of eligible households, Groundviews titled its investigation 'High Exclusion, Low Transparency', and policy institute LIRNEasia has called for clearer public explanation of how the algorithm works; reporting notes public blame for selection errors is often directed at 'the computer software' rather than at policy design.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Long (> 3 years)
Staffing & skills
Department of Census and Statistics enumerators and analysts, Welfare Benefits Board, Ministry of Finance, Planning and Economic Development

Conditions for success

  • Accurate, up-to-date household survey data
  • Transparent public communication of how the scoring algorithm works

Common failure modes

  • Arbitrary cutoffs and high exclusion of eligible households
  • Low transparency in how the algorithm operates
  • Public blame for errors misdirected at 'the software' rather than policy design

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful