Means-Tested vs Universal Approaches to Poverty: Comprehensive Study Notes
Key Concepts, Metrics and Terminology
Selective benefits
Cash transfers or services limited to individuals/households with limited resources.
May be delivered through direct means-testing or by targeting via other criteria (e.g. deprived area).
Means-tested benefit
Sub-category of selective benefits; entitlement requires assessment of income, assets, or both.
Universal benefits
Transfers/services available to all citizens / broad demographic groups without means-testing.
Can still be restricted by age, contribution history, etc. (e.g. state pension) but not by current income.
Contribution-based benefits
Universal in coverage; eligibility depends on prior tax/National Insurance contributions.
Earnings-related benefits
Level of benefit linked to previous earnings; horizontal redistribution dominates.
Targeting
Any method that selects beneficiaries (age, geography, income, family type, etc.).
Can occur inside otherwise “universal” programmes.
Redistributive impact
Change in the distribution of household income due to taxes and transfers.
Distinguish:
Horizontal (lifetime income-smoothing).
Vertical (from richer to poorer groups).
Size of redistributive budget
Measured as % of GDP spent on social programmes OR % of average disposable income comprised of transfers/services.
Progressivity
Extent to which net tax/benefit flows are higher (relative to income) for low-income groups.
Metrics:
Group-specific net transfers.
Concentration coefficient (analogous to Gini; lower values ⇒ more resources flow to the poor).
Standard formula for redistribution
\text{Redistribution} = \frac{Gini{market} - Gini{disposable}}{Gini_{market}}
Universalism vs. Selectivism: Conceptual Overview
All advanced welfare states combine universal and means-tested schemes; the balance varies by regime.
High-level implications for income redistribution and poverty, yet causal effects are complex and debated.
Historical debate: which approach better reduces poverty?
Both possess technical, behavioural and political drawbacks.
Limitations of Means-Tested Schemes
Stigma & Low Take-Up
Claimants must demonstrate need ⇒ perceived social failure.
2009/10 UK data: only 64\% of entitled people received Working Tax Credit; 62–69\% claimed Council Tax Benefit.
Administrative Costs
Means-testing is expensive, absorbing resources that could otherwise be paid out.
Inclusion / Exclusion Errors
Impossible to eliminate leakage to non-poor and exclusion of the “truly needy.”
Behavioural Distortions
Savings disincentive for pensions theoretically present but limited by uncertainty over future rules.
Poverty traps via high effective marginal tax rates (EMTRs):
Withdrawal of benefit (taper), start of income tax/NI contributions, loss of passported benefits (housing, prescriptions, education fees).
EMTR can approach 100\% or exceed it once work-related costs (e.g.
childcare) are included.
Political Vulnerability
Narrow beneficiary base ⇒ weak middle-class support; easier to cut or under-fund.
Sen’s maxim: “benefits meant exclusively for the poor often end up being poor benefits.”
Limitations of Universal Schemes
Fiscal Cost
Broad coverage drives high social-spending-to-GDP ratios; critics link this to reduced competitiveness.
Target Inefficiency
To reach a small poor population, government must also subsidise many non-poor households (Barry, 1990).
Distribution to Better-Off
Earnings-related transfers favour higher lifetime earners; horizontal redistribution dominates.
Differential life expectancy diminishes redistributive power of flat-rate pensions: low-income groups draw benefits for fewer years.
Political Risk of Austerity
Large, visible budgets tempt governments pursuing welfare retrenchment.
Political Economy and Public Support Dynamics
Means-testing may fracture coalitions between middle and working classes (Korpi & Palme, 1998).
Yet universal programmes can also be slashed when austerity bites (Pierson, 1994).
Voter motivations blend self-interest & moral sentiment; political context shapes outcomes (Rothstein, 2001).
Recent trend: even left-leaning parties embrace selective tools under fiscal pressure.
Empirical Evidence on Redistribution & Poverty
Early “Paradox of Redistribution” (Korpi & Palme, 1998)
Sample: 18 OECD states; LIS data for 11.
Findings:
Larger redistributive budgets correlate strongly with greater inequality reduction.
Greater targeting correlated with smaller budgets ⇒ less redistribution.
Countries with extensive earnings-related pensions (Sweden, Netherlands) showed lowest elder inequality despite weaker within-programme redistribution.
Core paradox: “the more we target benefits at the poor only … the less likely we are to reduce poverty and inequality.”
Mechanisms:
Inclusive schemes build cross-class coalitions ⇒ higher budgets.
Small, poor-only benefits discourage middle/upper classes; encourage private alternatives ⇒ higher overall inequality.
Studies Reinforcing Spending–Poverty Link
Cantillon et al. (2003): strong negative correlation between social expenditure (% GDP) and working-age poverty across 12 OECD nations.
Smeeding (2005): higher spending ⇒ lower poverty in 8 affluent countries.
Nolan & Marx (2009): link called “one of the most robust findings” in comparative research.
Brady & Barrow (2012): universal policies more effective for single-mother poverty than targeted policies.
Resilience to Retrenchment
Nelson (2007): 1990–2002, both social insurance & means-tested benefits eroded, but insurance benefits proved slightly more resilient; old-age pensions most protected.
Caveats: ignores coverage expansion of other targeted benefits (housing, tax credits).
Newer Evidence Challenging the Paradox
Whiteford (2008; 28 OECD countries)
Negative relation between size of budget and progressivity.
All cash systems progressive; highest progressivity: Australia, New Zealand, Denmark, UK (heavy means-testing).
Bottom quintile shares: Australia, Denmark, NZ, UK transfer largest shares; Scandinavian taxes claw back some gains.
Kenworthy (2011)
Re-runs Korpi & Palme for 1980-2005: positive universality–redistribution link fades after 1995.
Degree of universality did not predict changes in generosity under austerity.
Marx, Salanauskaite & Verbist (2013; 25 OECD)
No clear correlation between targeting and redistributive impact after isolating safety-net benefits.
Social assistance (means-tested) delivers largest progressivity share.
Positive universality–generosity link present but weaker than before.
Caminada et al. (2010)
Using net social spending and controls (unemployment, demographics, GDP pc) across 28 countries, higher spending did not systematically cut poverty.
Possible Explanations for Shift
Earlier samples small; addition of Southern Europe (large spending, weak targeting) alters relationships.
Policy design changes: stricter conditionality + in-work extension (e.g. Earned Income Tax Credit) improve means-tested reputation & funding.
Taxes matter: ignoring them overstated universalism’s redistributive edge.
United Kingdom in Comparative Perspective
Size of Redistributive Budget
Cash benefits ≈ 14.5\% of average household disposable income vs OECD mean 21.9\% (Whiteford, 2008).
Marx et al.: UK among lowest social transfers as % gross income.
Progressivity & Targeting
UK tax–benefit system labelled one of the most progressive (Whiteford, 2008).
31\% of all transfers flow to bottom quintile (vs Germany’s 28 % of disposable income with lower targeting).
Pre-transfer poor receiving any cash transfer: 97.6\% (Maitre et al., 2005).
ONS Table 1 (2011/12):
Market income ratio non-poor : poor ≈ 7:1; after redistribution 3:1.
Non-contributory transfers give poor households £$1{,}889$ more on average than non-poor.
Direct taxes: non-poor pay 7× poor in absolute £.
Composition of UK Transfers (ONS Figures)
Clearly pro-poor, means-tested programmes
Housing Benefit & Tax Credits each >10\% of poor household disposable income.
Income Support/Pension Credit, income-based JSA also strongly progressive.
Non income-tested programmes
Child Benefit & State Pension: similar absolute amounts across quintiles ⇒ mildly progressive due to income shares.
Incapacity Benefit: nearly 2× higher average payout to poor ⇒ reflects non-working status.
In-Kind Services
NHS & education overall progressive (Sefton, 2002; confirmed by ONS 2011/12).
NHS utilisation peaks in 2nd–3rd quintiles; education spending tilted to bottom quintile.
Poverty Reduction Performance
Pre-tax-transfer UK relative poverty near OECD average.
Smeeding (2005) ↓ from 31.8\% to 12.3\% (50 % median line).
OECD (2013) ↓ from 35.4\% to 17.2\% (60 % median line) – roughly halved.
Transfers lift 43 % of poor households above poverty line (Maitre et al., 2005).
Better than Southern Europe & Germany; below Scandinavia.
Summary of Main Findings (from Literature Review)
Recent evidence disputes the once-robust link between universalism, large budgets, and greater poverty reduction; the relationship is now ambiguous.
Country performance differs widely: Australia (high targeting) & Denmark (high universalism) both excel, indicating multiple viable models.
Structure and targeting of specific transfers (especially social assistance) matter as much as headline spending levels.
UK combines below-average spending with above-average targeting & progressive taxes, achieving mid-level poverty reduction.
Programmes such as Housing Benefit and Tax Credits are pivotal; without them inequality would widen sharply.
Policy implication: advocating more spending without design reform may disappoint; enhancing the reach and adequacy of well-targeted benefits, while safeguarding universality where politically and functionally justified, appears crucial.