
Georgia’s Targeted Social Assistance programme reaches more than 700,000 people — around one in five residents, including nearly a third of all children — even though official absolute poverty fell to 7.1% in 2025. At the same time, the average allowance per family remained below the family subsistence minimum. This analysis examines this apparent paradox and the programme’s interaction with the labour market, including rules that allow families to keep their status while in public works.
Key Findings
- In 2025, 397,100 families were registered in the Targeted Social Assistance database and 129,500 received the subsistence allowance; a further 54,800 families took part in the state public employment programme, which allows them to retain their socially vulnerable status (Geostat, 2026).
- In October 2025, 708,245 people received the subsistence allowance, at a monthly cost above GEL 71 million; in August 2025 nearly 279,000 recipients were children — close to a third of all children in Georgia (Georgia Today, 2025).
- Eligibility depends on a proxy means-tested rating score; families with scores up to 120,000 qualify. A score band of 100,001–120,000 added after 2020 contributed to the growth in recipients (PMCG, via Georgia Today).
- The average monthly allowance per family rose from GEL 163 in 2018 to GEL 256 in 2022, but remained below the subsistence minimum for an average family (GEL 314) (PMCG, via Georgia Today).
- In June 2026 the Prime Minister announced that a database review had terminated assistance for 22,707 families — 56.4% of those verified — saving GEL 54.7 million (Georgia Today, 2026).
1. Introduction
Targeted Social Assistance (TSA) is Georgia’s main cash benefit for poor households. Families apply to the Social Service Agency, which assigns a rating score based on household characteristics, income and assets; families below a threshold receive a subsistence allowance and additional benefits, including a child benefit. This analysis asks two questions: why does TSA cover far more people than official poverty statistics suggest are poor, and how does it interact with work?
2. Data
This analysis uses Geostat’s annual Indicators of Social Protection, which report registered and recipient families, and administrative data reported in the media, including recipient numbers by month. Absolute poverty figures are from Geostat. Monthly person counts and annual family counts are not directly comparable.
3. Findings
3.1 Scale and stability
The number of registered families rose from 390,800 in 2024 to 397,100 in 2025, and recipient families from 128,600 to 129,500 (Geostat, 2025; 2026). Tbilisi has the largest number: 79,600 registered families, of which 33,500 receive the allowance.
3.2 More recipients than poor people
| Indicator | Value |
|---|---|
| Subsistence allowance recipients, October 2025 | 708,245 people |
| Share of population (≈3.69 million) | ≈19% (GILS calculation) |
| Absolute poverty rate, 2025 | 7.1% |
| Child recipients, August 2025 | ≈279,000 (close to a third of children) |
Sources: Georgia Today (2025); Geostat. Share of population = GILS calculation.
Several factors explain why recipients outnumber people counted as poor. The rating threshold of 120,000 extends eligibility above the poverty line; families with a score up to 120,000 and children under 16 receive an increased child allowance, raised to GEL 150 per month from June 2022 (Ministry of Finance, 2023); child benefits create strong incentives for families with children to remain registered; and the public employment programme allows participants to keep their status for an extended period (Georgia Today, 2025). Poverty statistics are also measured after transfers: part of the fall in poverty may itself reflect TSA payments.
3.3 Allowances below subsistence
4. Discussion
Argument 1: The programme is doing two jobs
TSA functions both as poverty relief and, in practice, as Georgia’s main child benefit and a substitute for unemployment insurance. Georgia has no contributory unemployment benefit, so for many working-age adults without stable work, TSA is the only income support available. That helps explain its breadth, and why proposals to tighten eligibility meet resistance.
Argument 2: The work-incentive problem
Means-tested benefits that are withdrawn when income rises can discourage formal work: a family whose rating score crosses the threshold may lose both the allowance and the child benefit. The public employment programme, which preserves status while participants work, partly addresses this, but mainly through municipal public works (see GILS analysis of the Employment Promotion Law). A gradual taper, rather than a cliff, would reduce the penalty on formal employment.
Argument 3: Verification is necessary — and must be fair
The government’s 2026 review, which ended assistance for 22,707 families and cited recipients with monthly incomes of GEL 3,000–4,000 (Georgia Today, 2026), shows that eligibility checks matter. But large-scale terminations also carry risks of error for families whose circumstances are poorly captured by administrative data. Transparent criteria, notification and an accessible appeal process are essential.
Counter-argument: broad coverage is a strength
Some argue that broad coverage, especially of children, is a deliberate and effective way to reduce child poverty and should not be seen as a failure of targeting. GILS agrees that child support has value. The case for reform is not to cut coverage but to separate functions: a clear child benefit, a means-tested minimum income, and unemployment insurance linked to work.
5. Limitations
Recipient counts vary by month and source, and annual family counts cannot be converted directly into persons. The allowance-adequacy comparison refers to 2018–2022. This analysis did not assess the accuracy of the rating methodology.
6. Conclusion
Targeted Social Assistance reaches about one in five Georgians while paying average benefits below the subsistence minimum — broad but thin. It serves as poverty relief, child benefit and de facto unemployment support at once. Separating these functions, smoothing the benefit cliff and ensuring fair verification would make it both more effective and more compatible with work.
GILS Recommendations
- Introduce a gradual benefit taper so that families do not lose all support when earnings rise above the threshold.
- Separate the child benefit from the subsistence allowance, so that children’s support does not depend on parents remaining outside formal work.
- Introduce unemployment insurance, reducing reliance on TSA as income support for working-age adults.
- Guarantee due process in verification: notification, reasons and a timely appeal for families whose assistance is terminated.
- Publish an annual TSA evaluation, including targeting accuracy and labour market outcomes of recipients.
GILS Position
Georgia’s social assistance system protects many people, but often inadequately and in ways that can penalise work. GILS supports reform that keeps protection broad for children while making support for adults work-compatible and fair.
References
- Geostat (2026). Indicators of Social Protection (2025). geostat.ge
- Geostat (2025). Indicators of Social Protection 2024. geostat.ge
- GBC (2026). 888 thousand pensioners and about 130 thousand families are receiving social allowances — Geostat. gbc.ge
- Georgia Today (2025). Why Georgia has more welfare recipients than “poor” people — and what’s really behind it. georgiatoday.ge
- Georgia Today (n.d.). PMCG: Beneficiaries of the Subsistence Allowance. georgiatoday.ge
- Georgia Today (2026). Kobakhidze: social aid cut for 22,707 families after database review. georgiatoday.ge
- Ministry of Finance of Georgia (2023). Program Budget annex: social security programmes.
Suggested citation: GILS Research Team (2026). Georgia’s Social Assistance: Broad but Thin. Editor: G. Pkhakadze. Georgian Institute of Labour Studies, Tbilisi. labour.edu.ge
GILS analyses are prepared with AI-assisted drafting. All figures and claims are verified against the cited primary sources and reviewed under the responsibility of the Editor.