Mind the Gap: Coverage, Adequacy and Financing Gaps in Social Protection for the Extreme Poor and the Poorest Quintile

This paper evaluates the progress of social protection (SP) systems in Emerging and Developing Economies (EDEs), focusing on coverage, benefit adequacy, financing levels, and the targeting of social assistance, particularly for the extreme poor and the poorest quintile. It utilizes the World Bank’s ASPIRE database (combining household surveys and administrative data) to quantify gaps and explore policy options towards achieving Universal Social Protection (USP) and Sustainable Development Goal 1 (ending poverty). The analysis provides critical insights for policymakers aiming to strengthen SP systems, especially in resource-constrained environments.

Core Arguments & Findings

The Scale of the Coverage Gap: The Missing 2 Billion

Using household-level data (circa 2022), the paper identifies a significant gap in social protection within 130 Low- and Middle-Income Countries (LICs and MICs).

• 2 billion people remain either entirely missed by SP systems or receive inadequate support despite being poor (p. 20).
  • 1.6 billion live in households with no connection to SP (neither receiving benefits nor contributing to social insurance) (p. 21).
  • 0.4 billion are in the poorest quintile and receive benefits considered inadequate (below 20% of the relevant poverty line) (p. 18, p. 21).
• This gap persists despite noteworthy progress over the past decade (p. 3). The majority of these individuals reside in Sub-Saharan Africa, South Asia, and East Asia (p. 22).
• While the share of the population missed is highest in LICs (77%), the absolute number is larger in MICs (1.2 billion vs 500 million in LICs) due to population distribution (p. 21-22).

Gaps are Largest for the Poorest

Consistent with the principle of progressive realization of USP, the paper highlights that coverage gaps are most severe for the most vulnerable:

• Poorest Quintile: 53% of the 1.26 billion people in the poorest quintile across LICs and MICs are missed (25%) or inadequately covered (28%) (p. 24). Gaps are largest in LICs (97%) and Sub-Saharan Africa (94%) (p. 24).
• Extreme Poor: In 73 EDEs with significant extreme poverty (>2%), 87% of the extreme poor (approx. 430 million people) are missed or inadequately covered (p. 25). This rises to 98% in LICs (p. 26).
• In contrast, ‘new’ High-Income Countries (HICs) have largely closed these gaps, demonstrating that USP is achievable (p. 27).

Progress Made, But Pace is Slow

Cross-country analysis (using a narrower definition focusing on benefit receipt for 73 EDEs) shows progress over the decade preceding the COVID-19 shock (circa 2010 vs circa 2022):

• Average SP benefit coverage increased from 41% to 51% (p. 33).
• Progress occurred across all income groups but was most pronounced in LICs, where coverage doubled from a low base (11% to 25%) (p. 33).
• Coverage for the poorest quintile rose from 54% to 65%, and for the extreme poor from 59% to 70% (p. 33).
• Social assistance was the primary driver of this expansion, particularly for the poorest (p. 34-35). Social insurance coverage saw little growth (p. 36).
• However, at the current pace, achieving full coverage for the poor would take several decades, far beyond the SDG target date (p. 34).

Variable and Often Insufficient Benefit Adequacy

Benefit levels vary significantly by program type and country context:

• Social insurance programs (like pensions) offer the highest adequacy on average (replacing 33% of beneficiary welfare), while social assistance and labor market programs offer much less (10-11%) (p. 38).
• Adequacy ratios appear higher for the extreme poor and poorest quintile, but this is largely because their baseline consumption/income is very low (p. 38-39).
• While SP transfers reduce the pre-transfer income shortfall for recipients, the contribution is often insufficient, especially in high-poverty contexts and LICs (p. 41-43). On average, SP reduces the extreme poverty gap by only 10 percentage points for recipients (p. 42) and the poorest quintile gap by 16 percentage points (p. 43).

Significant Spending, But Major Financing Gaps

Social protection represents a notable area of government spending, but levels are often inadequate to fill coverage and adequacy gaps, particularly in LICs.

• EDEs spent an average of 5.3% of GDP on SP in 2022 (72 countries, admin data) (p. 45).
• Spending is strongly correlated with economic development: HICs spend 5.2 times more % GDP than LICs (p. 45). Absolute differences are starker: HICs spend US$2,608PPPpercapitavs.US$13 in LICs (151 times more) (p. 47).
• Social insurance dominates spending (3.7% GDP), followed by social assistance (1.5%) and labor market programs (0.2%) (p. 46). LICs rely more heavily on SA spending (p. 46).
• Low spending, often linked to low fiscal capacity, directly correlates with lower coverage and adequacy for the poor (p. 51-54).

Moderate Targeting Efficiency with Room for Improvement

The allocation of SP benefits shows a moderate pro-poor leaning overall, driven mainly by social assistance, but significant improvements are possible.

• Only 25% of overall SP beneficiaries are in the poorest quintile (Q1) (p. 55).
• Social assistance performs better, with 30% of beneficiaries in Q1 on average (54% in the bottom 40%). However, 27% are in the richest 40%, indicating inefficiency (p. 56).
• Targeting is weaker in LICs (27% SA beneficiaries in Q1) compared to UMICs (34%) (p. 56).
• The ‘pro-poor targeting frontier’ (performance of the top 10% best-targeted SA programs) shows that reaching 55-75% of beneficiaries in Q1 is feasible depending on country income level (p. 58-59). This suggests most EDEs could significantly improve targeting.

Key Statistics & Data

• Overall Gap (LICs/MICs, c. 2022): 2 billion people missed or inadequately covered (1.6B missed entirely, 0.4B poor receiving inadequate benefits) (p. 20-21).
• Coverage Increase (EDEs, c. 2010-2022): Average benefit receipt rose from 41% to 51% (p. 33).
• Coverage by Income (c. 2022): LICs 25%, LMICs 42%, UMICs 61%, New HICs 82% (p. 29).
• Poorest Quintile Gap (LICs/MICs): 53% missed or inadequately covered (p. 24).
• Extreme Poor Gap (High Poverty EDEs): 87% missed or inadequately covered (p. 25).
• LIC Extreme Poor Gap: 98% missed or inadequately covered (p. 26).
• Average SP Spending (EDEs, 2022): 5.3% of GDP (p. 45).
• Per Capita SP Spending (PPP, 2022): Average $1,002;HICs$2,608 vs LICs $13 (p. 47).
• SA Targeting (Average): 30% of beneficiaries in Q1 (p. 56).
• SA Targeting Frontier (Best Performers): 35-41% (LICs) to 63-88% (HICs) beneficiaries in Q1 (p. 59).
• Cost to Eradicate Extreme Poverty (EDEs, 2022): Additional US$137 billion/year (2017 PPP) (p. 64).
• Cost to Eliminate Poorest Quintile Shortfall (EDEs, 2022): Additional US$581 billion/year (2017 PPP) (p. 64).

Methodology

The paper relies primarily on the World Bank’s ASPIRE (Atlas of Social Protection: Indicators of Resilience and Equity) database, which compiles:

• Nationally representative household survey data: Used to calculate coverage (direct/indirect beneficiaries, sometimes contributors), adequacy ratios, targeting incidence (quintile distribution), and poverty impacts. Coverage estimates in Section IIA use a comprehensive definition (beneficiaries + contributors), while the rest of the paper focuses on benefit receipt (p. 9, 17, 28). Data circa 2022 (mostly 2018-2022 surveys) and circa 2010 (2006-2014 surveys) are used (p. 10, 19, 33).
• Administrative program-level data: Used to track SP expenditure (as % GDP and per capita PPP) and composition across 72 countries for 2017-2022 (p. 9, 44, 82).
• Poverty Gap Analysis: Calculates the income shortfall needed to lift individuals/households above the extreme poverty line ($2.15/day PPP) or the relative poverty line (poorest quintile threshold) before and after SP transfers (p. 42, 62).
• Targeting Analysis: Assesses the distribution of program beneficiaries across consumption/income quintiles (p. 56). Defines a ‘pro-poor targeting frontier’ based on best-performing programs (p. 59).
• Simulation: Estimates the potential reduction in poverty gaps achievable by improving the pro-poor efficiency of existing social assistance spending, based on feasible targeting parameters (derived from the frontier) and assumptions about program mix and administrative costs (Annex 2, p. 60-61, 86-89).

Limitations acknowledged: Include lags in household survey data availability for some countries, potential underestimation of total SP coverage/spending if surveys/admin data miss programs, and reliance on assumptions for estimations and simulations (p. 19, 87).

Key Conclusions & Recommendations

Conclusions

1. Significant Gaps Persist: Despite progress, 2 billion people in LICs/MICs lack adequate social protection, with extreme coverage gaps (over 85%) for the poor in LICs and high-poverty settings (p. 20, 26).
2. Adequacy is Highly Variable: Benefits, especially from social assistance, are often too low to significantly reduce poverty or protect against shocks, particularly in LICs (p. 43).
3. Targeting Can Be Much More Efficient: Social assistance reaches the poor better than other SP areas, but substantial leakage to non-poor households exists. Significant efficiency gains are technically feasible across all country income levels by adopting practices of best-performing programs (p. 56, 59).
4. Financing is Insufficient and Mismatched: While global SA spending is theoretically large, resources are concentrated in higher-income countries. LICs face severe financing constraints, limiting their ability to expand coverage and adequacy (p. 45, 63, 71).
5. Closing Gaps Requires Multiple Strategies: Improving efficiency alone is insufficient, especially for eliminating extreme poverty in LICs. Additional resources (domestic and international) are crucial (p. 65, 74).

Recommendations (Stated or Implied)

1. Prioritize the Poorest (Progressive Universalism): Focus initial efforts on closing coverage and adequacy gaps for the extreme poor and poorest quintile, particularly through social assistance (p. 8, 14,