## Annex I. Prerequisites for Effective Targeting Systems

## Source details

**Canonical URL:** [Annex I. Prerequisites for Effective Targeting Systems](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf.json)

---

### Overview
- The Covid-19 pandemic exposed significant weaknesses of social safety net (SSN) systems in emerging and developing economies (EMDEs), including gaps in coverage of poor households and subsequent scaling back of pandemic-era expansions (Gentilini, 2023).
- Strengthening SSNs requires evaluating systems in terms of both spending efficiency and spending adequacy using a social welfare–based framework.
- Key policy challenges identified:
  - large cross-country differences in SSN effectiveness driven largely by spending adequacy rather than targeting efficiency;
  - weak capacity to channel a high share of SSN spending to lower-income households;
  - a trade-off between high coverage of the poor and containment of fiscal cost;
  - tight short-term fiscal constraints; and
  - the potential role of proxy-means targeting.

### Prerequisites for Effective Targeting Systems (summarized findings and implications)
- Capacity to channel spending to lower-income households
  - Weak capacity to channel (or target) a high share of SSN spending to lower-income households limits ability to address low coverage of the poor without substantial leakage to higher-income households.
  - This limitation creates a trade-off between achieving high coverage of the poor and containing associated fiscal costs.
- Fiscal space and spending adequacy
  - Differences in SSN effectiveness across countries are driven in large part by differences in spending levels relative to the size of the poverty problem (i.e., adequacy).
  - Strengthening SSNs will therefore require increasing social assistance spending in many countries, especially low-income countries, alongside measures to create fiscal space.
- Administrative capacity and country capabilities
  - Tight fiscal constraints, competing claims on public spending, and weak capacity for increasing tax revenues mean improvements require not only fiscal space but also enhanced country capacities to effectively target SSN spending to low-income households.
  - Administrative and implementation capacity is implied as a prerequisite to improve targeting efficiency and to manage trade-offs between coverage and leakage.
- Use and design of proxy-means targeting
  - More widespread use of proxy-means targeting methods can help to strengthen targeting efficiency in the short term.
  - Proxy-means targeting needs careful design to reduce associated horizontal equities related to different treatment of equally poor households.
- Need for country-specific policy prioritization
  - Setting country-specific policy priorities should be informed by comprehensive evaluation of SSN systems’ effectiveness at reducing poverty in terms of spending efficiency and spending adequacy, using available cross-country information such as the World Bank’s ASPIRE database.

### Policy implications and directions (Annex I)
- Prioritize increasing social assistance spending where adequacy is the binding constraint, notably in low-income countries.
- Enhance administrative and targeting capacities to increase the share of SSN benefits reaching lower-income households and to better manage the coverage–cost trade-off.
- Consider carefully designed proxy-means targeting as a short-term measure to improve targeting efficiency while addressing horizontal equity concerns.
- Use the social welfare–based evaluation framework and cross-country databases (for example, ASPIRE) to inform country-specific strategies and to monitor performance in terms of both efficiency and adequacy.

### Section V: Trade-off between expanding poor coverage and transfer leakage — conceptual framework and data
- Conceptual framework and definitions
  - Final income: 푦1 = 푦0 + 푚.
  - Social welfare marginal impact: dW = ∑βh dmh where βh are social welfare weights.
  - Distributional characteristic (transfer efficiency): λ such that PI = λ.B (equation 5).
  - Special cases:
    - Perfect targeting: λ = 1 and PI equals the transfer budget B.
    - Imperfect targeting: λ < 1 due to leakage to non-poor and overcompensation of some poor households.
  - Behavioral responses (e.g., reduced labor supply) are abstracted from in this analysis (footnote 5).
- ASPIRE variables and relationships
  - Poverty Impact defined as the percentage change in the poverty gap index (∆PGI).
  - Benefit-Cost Ratio (BCR) = transfer efficiency (λ).
  - Coverage (COV): COV-P (percent of poor households receiving transfers) and COV-T (percent of all households receiving transfers).
  - Relationship used: ∆PGI = BCR · BR, where BR = transfer budget relative to pre-transfer total poverty gap (equation 6).
  - BR can be backed out as (∆PGI / BCR).
  - Total pre-transfer poverty gap PGpre = B / BCR, using B and BCR provided by ASPIRE team.

### Section V empirical findings and statistics
- Main empirical findings:
  - For the 20% poverty line, variation in transfer budget ratio (BR) explains 84 percent of total variation in poverty impact; variation in efficiency (BCR) explains 19 percent.
  - ANOVA decomposition: variation in budget ratios explains just over 90 percent of the total explained variation for the 20% poverty line.
  - For extreme poverty, ANOVA shows BR and BCR each explain approximately half of the total explained variation.
- Regional observations (20% poverty gap, selected values)
  - EDA and MENA: PGIs of 0.059.
  - LAC: PGI of 0.085.
  - CCA, EMDE, SSA: PGIs approximately 0.073.
- Sample sizes referenced
  - Figure 2 sample: 79 countries (CCA(7), EMDE(12), LAC(15), EDA(16), MENA(5), SSA(24)).
  - Figures 3–4 sample: 59 countries (CCA(4), EDE(8), LAC(13), EDA(13), MENA(2), SSA(19)).
- Interpretation
  - Across EMDEs, the primary driver of differences in poverty impact is budget effort (B), particularly for the 20% poverty line and for low-income countries under extreme poverty.

### Household coverage and transfer efficiency (empirical relationships and regression results)
- Core trade-off: expanding coverage of poor households versus containing leakage to non-poor households.
- Administrative and structural constraints limiting means-testing
  - Low administrative capacity, large informal sector, multiple/volatile income sources, poor bookkeeping, political/social reluctance to means-test.
- Categorical targeting trade-offs
  - Narrow categorical targeting: high transfer concentration to low-income households but large coverage gaps (high undercoverage).
  - Broad categorical targeting: expands coverage but increases leakage to higher-income households.
- Empirical relationship between total coverage (COV-T) and poor coverage (COV-P)
  - Strong positive relationship: “no country achieves high coverage of poor households without having high total coverage.”
  - All countries with coverage of the poor above 80 percent have total population coverage above 50 percent.
  - For the poorest 20 percent, near full coverage of the poor requires total coverage above 70 percent.
- Regression table (selected coefficients and statistics reproduced exactly)
  - Total coverage coefficient rows: 1.125***, 1.906***, 1.069***, 2.219***, 0.957***, 1.879***.
  - Total coverage squared row: -0.928***, -1.356***, -1.072***.
  - Poverty Headcount coefficient row: -0.634***, -0.567***.
  - Constants: 0.124***, 0.189***, 0.199***, 0.040***, 0.341***, 0.200***.
  - R-squared: 0.872, 0.906, 0.688, 0.758, 0.807, 0.85.
  - Number of observations: 70, 70, 61, 61, 60, 60.
- Simulation results (required total coverage to hit poor-coverage targets)
  - To reach COV-P of (80, 90, 100) percent for the poorest 20 percent, required total coverage COV-T ≈ (60, 69, 78) percent (ignoring H).
  - Corresponding required total coverage for extreme poverty: (56, 66, 75) percent (ignoring H).

### Transfer spending adequacy (Section VI findings and illustrative calculation)
- Minimum budget under perfect targeting = total pre-transfer poverty gap (PGpre).
- Empirical patterns
  - Minimum budget gap > 1 percent of GDP in many countries, especially in SSA.
  - Outside SSA, the minimum budget gap is lower or even negative in some EDE, CCA, and LAC countries.
- Accounting for leakage when expanding coverage
  - Conservative target example: to achieve 90 percent coverage of the poor under the 20% poverty line, one must cover 70 percent of the population rather than 20 percent under the minimum budget assumption.
  - Under a uniform transfer at the average individual poverty gap, required budget with leakage = 70/20 = 3.5 times the minimum budget.
  - Result: required budget increase to achieve this exceeds 3 percent of GDP in many countries, especially in SSA.
- Fiscal realism
  - Many EMDEs have total tax ratios below 15 percent of GDP and competing development spending claims, making large budget increases infeasible in the short term.
  - Many countries would struggle to raise government revenues by 1 percentage point of GDP in the short term.

### Enhancing targeting efficiency and proxy-means tests (Section VII)
- Short-term priority: enhance targeting efficiency (BCR) so limited budgets reach poor households more effectively.
- Proxy-means tests (PMTs)
  - PMT is a scoring method using household characteristics correlated with welfare.
  - Simulations (Coady and Le, 2020; India example) show PMT can substantially reduce the coverage–leakage trade-off:
    - PMT simulated coverage: covers 50 percent of the population with ~90 percent coverage of bottom two deciles; coverage falls to ~50 percent at the fifth decile and almost zero at the top decile.
    - Compared to categorical targeting empirical relationship (Figure 5), a PMT could imply a lower budget requirement at about 70 percent (i.e., 50/70) of that estimated under broad categorical targeting.
  - PMT caveats: risk of leakage, undercoverage, horizontal inequity, community unrest, and poor responsiveness to income shocks.
- PMT design and implementation steps (practical, prioritized actions)
  - Prioritize unifying beneficiary databases into a unified registry to eliminate double-dipping and cross-check information (example: Brazil’s Cadastro Unico description).
  - Increase coverage with simple progressive demographic-based transfers (e.g., child benefits 0–5 years).
  - Combine universal categorical programs with self-selection programs (e.g., geographically targeted public works) and differentiate benefits across programs.
  - Adopt application-based initial PMT screening (hybrid PMTs), with careful attention to affordability of opt-in for the poor and verification capacity (example: Moldova’s Ajutor Social hybrid two-stage approach).
  - Provide appeals procedures for households deemed ineligible by basic PMTs, with transparent overturn criteria and potential community involvement.
  - Gradually expand PMT programs starting with the poorest municipalities and complement with universal categorical programs as administrative capacity scales up (example: Peru’s Juntos three-step process).
  - Exclude high-income households using social security income data and claw back benefits via the personal income tax system where feasible.
- Additional PMT design points
  - Differentiating benefit levels by proxy-means score increases the share of transfers to lower-income groups above their coverage share and reduces horizontal inequity at the margin.
  - Hybrid PMTs and staged implementation can mitigate some misclassification and verification problems.

### Summary and policy implications (Section VIII conclusions)
- Framework and decomposition
  - Poverty impact PI decomposed into transfer efficiency (λ = BCR) and budget effort (B).
- Main conclusions from ASPIRE-based analysis
  - Variation in SSN poverty alleviation across EMDEs is largely driven by variation in SSN budgets (especially for 20% poverty line and low-income countries for extreme poverty).
  - No country in the EMDE sample achieves high coverage of poor households without also having high total population coverage, implying substantial leakage when expanding coverage with categorical methods.
  - Required spending increases that account for inevitable leakage (to achieve high poor coverage) are substantially larger than the minimum budget gap and are prohibitive for most low-income countries.
- Policy priorities given fiscal constraints
  - Increase SSN budgets where feasible, but recognize short-term revenue constraints (difficulty raising tax ratios and competing development claims).
  - Improve transfer efficiency as a practical short-term strategy to increase reach to the poor while containing leakage.
  - Carefully designed PMTs (including hybrid approaches, unified registries, phased rollouts, appeals mechanisms, and benefit differentiation) can reduce the required spending by about one third or more; this reduction can be larger if transfer levels are differentiated by proxy-means scores.
- Implementation caveats
  - PMTs require complementary investments (information systems, administrative capacity, institutional arrangements, fiscal capacity).
  - Horizontal inequity, community acceptance, and the capacity to verify and respond to shocks must be managed through design and implementation choices.

*Source: IMF Working Paper — Challenges Facing SSNs in Emerging and Developing Economies: An Illustration Using the World Bank’s ASPIRE Database (Annex I; Sections V–VIII). Canonical URL: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf*

### Annex I. Prerequisites for Effective Targeting Systems .................................................................

### Annex I. Prerequisites for Effective Targeting Systems

### Overview
- The Covid-19 pandemic exposed significant weaknesses of social safety net (SSN) systems in emerging and developing economies (EMDEs), including gaps in coverage of poor households and subsequent scaling back of pandemic-era expansions (Gentilini, 2023).
- Strengthening the effectiveness of SSNs requires evaluating systems in terms of both spending efficiency and spending adequacy using a social welfare–based framework.
- Key policy challenges identified include: large cross-country differences in SSN effectiveness driven largely by spending adequacy rather than targeting efficiency; weak capacity to channel a high share of SSN spending to lower-income households; a trade-off between high coverage of the poor and containment of fiscal cost; tight short-term fiscal constraints; and the potential role of proxy-means targeting.

### Prerequisites for Effective Targeting Systems (summarized findings and implications)
- Capacity to channel spending to lower-income households
  - Weak capacity to channel (or target) a high share of SSN spending to lower-income households is a key challenge limiting ability to address low coverage of the poor without substantial leakage of benefits to higher-income households.
  - This limitation creates a trade-off between achieving high coverage of the poor and containing associated fiscal costs.

- Fiscal space and spending adequacy
  - Differences in SSN effectiveness across countries are driven in large part by differences in spending levels relative to the size of the poverty problem (i.e., adequacy).
  - Strengthening SSNs will therefore require increasing social assistance spending in many countries, especially low-income countries, alongside measures to create fiscal space.

- Administrative capacity and country capabilities
  - In the short term, tight fiscal constraints, competing claims on public spending, and weak capacity for increasing tax revenues mean improvements require not only fiscal space but also enhanced country capacities to effectively target SSN spending to low-income households.
  - Administrative and implementation capacity is implied as a prerequisite to improve targeting efficiency and to manage trade-offs between coverage and leakage.

- Use and design of proxy-means targeting
  - More widespread use of proxy-means targeting methods can help to strengthen targeting efficiency in the short term.
  - Proxy-means targeting needs to be carefully designed to reduce associated horizontal equities related to different treatment of equally poor households.

- Need for country-specific policy prioritization
  - Setting country-specific policy priorities should be informed by comprehensive evaluation of SSN systems’ effectiveness at reducing poverty in terms of spending efficiency and spending adequacy, using available cross-country information such as the World Bank’s ASPIRE database.

### Policy implications and directions
- Prioritize increasing social assistance spending where adequacy is the binding constraint, notably in low-income countries.
- Enhance administrative and targeting capacities to increase the share of SSN benefits reaching lower-income households and to better manage the coverage–cost trade-off.
- Consider carefully designed proxy-means targeting as a short-term measure to improve targeting efficiency while addressing horizontal equity concerns.
- Use the social welfare–based evaluation framework and cross-country databases (for example, ASPIRE) to inform country-specific strategies and to monitor performance in terms of both efficiency and adequacy.

*Source: IMF Working Paper — Challenges Facing SSNs in Emerging and Developing Economies: An Illustration Using the World Bank’s ASPIRE Database (Annex I: Prerequisites for Effective Targeting Systems).*

### Section V examines the trade-off that exists between expanding coverage of poor households and transfer

### Section V examines the trade-off that exists between expanding coverage of poor households and transfer

### Conceptual framework: poverty impact as product of targeting efficiency and budget effort
- Final income: 푦1 = 푦0 + 푚.
- Social welfare marginal impact: dW = ∑βh dmh where βh are social welfare weights.
- Distributional characteristic (transfer efficiency): λ such that PI = λ.B (equation 5).
- Special cases:
  - Perfect targeting: λ = 1 and PI equals the transfer budget B.
  - Imperfect targeting: λ < 1 due to leakage to non-poor and overcompensation of some poor households.
- Notes on behavioral responses: behavioral responses (e.g., reduced labor supply) are abstracted from in this analysis (footnote 5).

### Data and metrics from the World Bank’s ASPIRE database
- Key ASPIRE variables used:
  - Poverty Impact defined as the percentage change in the poverty gap index (∆PGI).
  - Benefit-Cost Ratio (BCR) = transfer efficiency (λ).
  - Coverage (COV): COV-P (percent of poor households receiving transfers) and COV-T (percent of all households receiving transfers).
- Relationship used: ∆PGI = BCR · BR, where BR = transfer budget relative to pre-transfer total poverty gap (equation 6).
- Derivations:
  - BR can be backed out as (∆PGI / BCR).
  - Total pre-transfer poverty gap PGpre = B / BCR, using B and BCR provided by ASPIRE team.

### Findings on poverty impact, transfer efficiency, and budget effort
- Main empirical findings:
  - For the 20% poverty line, variation in transfer budget ratio (BR) explains 84 percent of total variation in poverty impact; variation in efficiency (BCR) explains 19 percent.
  - ANOVA decomposition: variation in budget ratios explains just over 90 percent of the total explained variation for the 20% poverty line.
  - For extreme poverty, ANOVA shows BR and BCR each explain approximately half of the total explained variation.
- Regional observations (20% poverty gap, selected values):
  - EDA and MENA: PGIs of 0.059.
  - LAC: PGI of 0.085.
  - CCA, EMDE, SSA: PGIs approximately 0.073.
- Sample sizes referenced:
  - Figure 2 sample: 79 countries (CCA(7), EMDE(12), LAC(15), EDA(16), MENA(5), SSA(24)).
  - Figures 3–4 sample: 59 countries (CCA(4), EDE(8), LAC(13), EDA(13), MENA(2), SSA(19)).
- Interpretation:
  - Across EMDEs, the primary driver of differences in poverty impact is budget effort (B), particularly for the 20% poverty line and for low-income countries under extreme poverty.

### Household coverage and transfer efficiency (Section V specific findings)
- Core trade-off: expanding coverage of poor households versus containing leakage to non-poor households.
- Administrative and structural constraints limiting means-testing: low administrative capacity, large informal sector, multiple/volatile income sources, poor bookkeeping, political/social reluctance to means-test (paragraph summary).
- Categorical targeting trade-offs:
  - Narrow categorical targeting: high transfer concentration to low-income households but large coverage gaps (high undercoverage).
  - Broad categorical targeting: expands coverage but increases leakage to higher-income households.
- Empirical relationship (Figure 5 and regressions):
  - Strong positive relationship between total population coverage (COV-T) and coverage of the poor (COV-P): “no country achieves high coverage of poor households without having high total coverage.”
  - All countries with coverage of the poor above 80 percent have total population coverage above 50 percent.
  - For the poorest 20 percent, near full coverage of the poor requires total coverage above 70 percent.
- Regression table (selected coefficients and statistics reproduced exactly):
  - Total coverage coefficient rows: 1.125***, 1.906***, 1.069***, 2.219***, 0.957***, 1.879***.
  - Total coverage squared row: -0.928***, -1.356***, -1.072***.
  - Poverty Headcount coefficient row: -0.634***, -0.567***.
  - Constants: 0.124***, 0.189***, 0.199***, 0.040***, 0.341***, 0.200***.
  - R-squared: 0.872, 0.906, 0.688, 0.758, 0.807, 0.85.
  - Number of observations: 70, 70, 61, 61, 60, 60.
- Simulation for required total coverage to hit poor-coverage targets (using regression):
  - To reach COV-P of (80, 90, 100) percent for the poorest 20 percent, required total coverage COV-T ≈ (60, 69, 78) percent (ignoring H).
  - Corresponding required total coverage for extreme poverty: (56, 66, 75) percent (ignoring H).

### Transfer spending adequacy (Section VI findings and illustrative calculation)
- Minimum budget under perfect targeting = total pre-transfer poverty gap (PGpre).
- Empirical patterns:
  - Minimum budget gap > 1 percent of GDP in many countries, especially in SSA.
  - Outside SSA, the minimum budget gap is lower or even negative in some EDE, CCA, and LAC countries.
- Accounting for leakage when expanding coverage:
  - Conservative target example: to achieve 90 percent coverage of the poor under the 20% poverty line, one must cover 70 percent of the population rather than 20 percent under the minimum budget assumption.
  - Under a uniform transfer at the average individual poverty gap, required budget with leakage = 70/20 = 3.5 times the minimum budget.
  - Result: required budget increase to achieve this exceeds 3 percent of GDP in many countries, especially in SSA.
- Fiscal realism:
  - Many EMDEs have total tax ratios below 15 percent of GDP and competing development spending claims, making large budget increases infeasible in the short term.
  - Many countries would struggle to raise government revenues by 1 percentage point of GDP in the short term.

### Enhancing targeting efficiency (Section VII policy recommendations and simulation evidence)
- Short-term priority: enhance targeting efficiency (BCR) so limited budgets reach poor households more effectively.
- Proxy-means tests (PMTs):
  - PMT is a scoring method using household characteristics correlated with welfare.
  - Simulations (Coady and Le, 2020; India example) show PMT can substantially reduce the coverage–leakage trade-off:
    - PMT simulated coverage: covers 50 percent of the population with ~90 percent coverage of bottom two deciles; coverage falls to ~50 percent at the fifth decile and almost zero at the top decile.
    - Compared to categorical targeting empirical relationship (Figure 5), a PMT could imply a lower budget requirement at about 70 percent (i.e., 50/70) of that estimated under broad categorical targeting.
  - PMT caveats: risk of leakage, undercoverage, horizontal inequity, community unrest, and poor responsiveness to income shocks.
- PMT design and implementation steps (practical, prioritized actions):
  - Prioritize unifying beneficiary databases into a unified registry to eliminate double-dipping and cross-check information (example: Brazil’s Cadastro Unico description).
  - Increase coverage with simple progressive demographic-based transfers (e.g., child benefits 0–5 years).
  - Combine universal categorical programs with self-selection programs (e.g., geographically targeted public works) and differentiate benefits across programs.
  - Adopt application-based initial PMT screening (hybrid PMTs), with careful attention to affordability of opt-in for the poor and verification capacity (example: Moldova’s Ajutor Social hybrid two-stage approach).
  - Provide appeals procedures for households deemed ineligible by basic PMTs, with transparent overturn criteria and potential community involvement.
  - Gradually expand PMT programs starting with the poorest municipalities and complement with universal categorical programs as administrative capacity scales up (example: Peru’s Juntos three-step process).
  - Exclude high-income households using social security income data and claw back benefits via the personal income tax system where feasible.
- Additional PMT design points:
  - Differentiating benefit levels by proxy-means score increases the share of transfers to lower-income groups above their coverage share and reduces horizontal inequity at the margin.
  - Hybrid PMTs and staged implementation can mitigate some misclassification and verification problems.

### Summary and policy implications (Section VIII conclusions)
- Framework and decomposition:
  - Poverty impact PI decomposed into transfer efficiency (λ = BCR) and budget effort (B).
- Main conclusions from ASPIRE-based analysis:
  - Variation in SSN poverty alleviation across EMDEs is largely driven by variation in SSN budgets (especially for 20% poverty line and low-income countries for extreme poverty).
  - No country in the EMDE sample achieves high coverage of poor households without also having high total population coverage, implying substantial leakage when expanding coverage with categorical methods.
  - Required spending increases that account for inevitable leakage (to achieve high poor coverage) are substantially larger than the minimum budget gap and are prohibitive for most low-income countries.
- Policy priorities given fiscal constraints:
  - Increase SSN budgets where feasible, but recognize short-term revenue constraints (difficulty raising tax ratios and competing development claims).
  - Improve transfer efficiency as a practical short-term strategy to increase reach to the poor while containing leakage.
  - Carefully designed PMTs (including hybrid approaches, unified registries, phased rollouts, appeals mechanisms, and benefit differentiation) can reduce the required spending by about one third or more; this reduction can be larger if transfer levels are differentiated by proxy-means scores.
- Implementation caveats:
  - PMTs require complementary investments (information systems, administrative capacity, institutional arrangements, fiscal capacity).
  - Horizontal inequity, community acceptance, and the capacity to verify and respond to shocks must be managed through design and implementation choices.

*Source: IMF Working Paper — Challenges Facing SSNs in Emerging and Developing Economies: An Illustration Using the World Bank’s ASPIRE Database (wpiea2024096-print-pdf). Canonical URL: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf*

### Annex I. Prerequisites for Effective Targeting

### Annex I. Prerequisites for Effective Targeting

### Systems — overview
- High-performing SSN systems that accomplish broad coverage of poor and vulnerable households require the capacity to collect and process a wide range of socio-economic information to enable cost-effective program expansion through efficient targeting.
- Collecting information beyond income is crucial given the difficulties in verifying income of poor and vulnerable households with multiple and volatile sources of income.
- Governments need the capacity to deliver transfers to eligible beneficiaries in a timely, reliable, and cost-effective manner.
- Enhancing targeting capacity requires investment in the requisite institutional infrastructure (Grosh and others, 2008, 2022; Brollo and others, 2024).

### Strong information systems
- Components:
  - Comprehensive and reliable identification systems.
  - Ability to collect and verify existing and new socio-economic information (e.g., data on household characteristics, employment, and income).
  - Capacity to store and manage this information in a timely manner.
- Evidence from the pandemic:
  - Countries with comprehensive integrated social registries containing up-to-date information and broad coverage of the poor and the vulnerable population were able to successfully leverage these systems to provide additional support to existing beneficiaries and expand coverage to include additional vulnerable households at reasonable fiscal cost.
  - Some countries lacking unified social registries but with well-integrated sources of administrative data (including registries of beneficiaries of different social protection programs, civil registries, and tax records) were able to leverage this information to scale-up SSN programs and verify eligibility.
  - Countries that lacked social registries and integrated (or “interoperable”) sources of administrative data, or where those systems had low coverage of the vulnerable population and out-of-date information, faced difficulties identifying those in need and had to collect new information under tight timeframes, with challenges for eligibility verification.

### Strong implementation capacity
- Components:
  - Capacity to deliver benefits to intended beneficiaries in a reliable, cost-effective, and timely manner.
  - Administrative systems for service delivery, integrated across social protection programs.
  - Robust systems for informed decision-making and accountability enhancement (e.g., grievance mechanisms and monitoring and evaluation systems).
  - Adequate human resource capabilities and staffing.
- Digital transfers:
  - Electronic payment systems (“digital transfers”) can help to increase the efficiency and transparency of cash-transfer delivery.
  - Digital payments require an adequate payment ecosystem, including:
    - A strong legal framework for digital payments.
    - Interoperable payments infrastructure.
    - Broad electronic payments acceptance.
    - Strong cash out networks.
    - Financial literacy.

### Strong institutional arrangements
- Components:
  - Capacity to develop clear and appropriate national social protection policies, implementation strategies, and supporting laws and regulations.
  - Strong government leadership and clear roles and responsibilities of the different government entities involved in social protection.
  - Strong coordination mechanisms to enhance efficiency and effectiveness of SSNs and their response to shocks by reducing duplication, avoiding unnecessary delays, and improving response to the needs of affected populations.
  - Where relevant, clear standards and procedures to guide the integration of non-governmental organizations and humanitarian actors.

### Strong fiscal systems
- Requirements:
  - Creating the needed fiscal space in a sustainable and progressive manner to avoid undermining fiscal and macroeconomic stability.
  - Predicated on credible Medium-term Fiscal Frameworks (MTFFs) that reflect other spending needs and revenue mobilization capacity, including the required fiscal and economic reforms.
- Benefits of strong MTFFs:
  - Help build fiscal buffers.
  - Can facilitate the withdrawal of discretionary fiscal measures as the economy recovers.

*Source: Annex I. Prerequisites for Effective Targeting, from Challenges Facing SSNs in Emerging and Developing Economies: An Illustration Using the World Bank’s ASPIRE Database, Working Paper No. WP/2024/096.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024096-print-pdf.pdf_