Formalising Public Choice Theory & Policy Targeting: Social Welfare, and Inequality

Formalising Public Choice Theory & Policy Targeting: Social Welfare, and Inequality

Overview of the Lecture Series

  • This lecture series focuses on public interventions in economic outcomes.

  • Key questions addressed:

    • When is government intervention in the economy appropriate?

    • How can the government intervene effectively?

    • What are the economic consequences of these interventions?

    • What factors drive government choices regarding intervention?

  • Topics covered include:

    • Social insurance

    • Social assistance and redistribution

    • The impacts of social grants

    • General social spending

Social Insurance

Definition and Characteristics

  • Social insurance programs: Government interventions designed to provide insurance against specified adverse events.

  • Eligibility: For most programs, eligibility is not means-tested.

    • Means-tested programs: Programs where eligibility is contingent on an individual's current income or asset levels.

The South African Social Security System (Overview)

  • The system comprises two main components: Social Insurance and Social Assistance.

  • Social Insurance:

    • Components: Unemployment Insurance Fund, Compensation Funds, Road Accident Fund.

    • Financing Mechanism: Payroll taxes and fuel levy.

    • Funding Sources: Workers, Employers, Road users.

  • Social Assistance:

    • Components: Child-support grant, Care-dependency grant, Foster-care grant, Disability grant, Old-age pension, War veteran's grant, Grant-in-aid.

    • Financing Mechanism: Other taxes.

    • Funding Sources: Workers, Employers, Other taxpayers.

  • Occupational Insurance (Private):

    • Components: Retirement funds, Medical schemes.

    • Financing Mechanism: Contributions.

    • Funding Sources: Workers, Employers.

Social Insurance in South Africa: Key Funds

  • There are three primary contributory funds:

    • Unemployment Insurance Fund (UIF):

      • Provides insurance against short-term unemployment due to retrenchment or illness.

      • Also offers maternity leave benefits.

      • Funded by contributions from both employers and employees.

    • Compensation Funds:

      • Employers make risk-related contributions.

      • Benefits are paid to employed workers who become temporarily or permanently disabled due to work-related incidents.

    • Road Accident Fund (RAF):

      • Provides cover to all road users within South Africa for injuries sustained or death arising from motor vehicle accidents.

  • Employer Pension/Provident Funds:

    • Based on conventional and binding agreements.

    • Primarily private, with government involvement being regulatory.

Occupational Retirement Insurance Statistics (South Africa)

  • There are approximately 9,505 retirement funds, covering 13.8 million members.

  • The disbursed amount from these funds exceeds social grant payments by more than 50%.

  • Coverage is not universal:

    • Limited to those in paid employment.

    • Coverage is notably low in sectors such as agriculture, trade, catering, accommodation, and domestic services.

    • Overall, only about 47% of the labour force is covered.

Market Failures in the Insurance Market

  • Adverse Selection:

    • This occurs due to hidden type problems.

    • Individuals most likely to experience a risk are also the most likely to purchase insurance.

    • Government intervention, such as an individual mandate, can potentially mitigate this problem.

  • Moral Hazard:

    • This arises from hidden action problems.

    • The presence of insurance can increase the likelihood of individuals engaging in risky behavior.

    • Governments typically find it challenging to solve moral hazard issues effectively.

Social Welfare & Inequality

Income Inequality in South Africa and Emerging Economies

  • Data from Leibbrandt et al. (2010) shows persistent income inequality in South Africa, with higher deciles consistently holding larger shares of total income across 1993, 2000, and 2008.

  • Compared to other emerging countries (1910-2010), South Africa historically exhibited and often maintained one of the highest shares of total income held by the top percentile, standing out amongst countries like Argentina, Colombia, Indonesia, India, and China.

Income Distribution and the Role of Social Assistance

  • The significant discrepancy in resources among citizens poses a central question for public finance.

  • Objective: Social welfare could be maximised by redistributing wealth from high-income to low-income individuals.

  • Private Sector Limitation: The private sector is unlikely to provide this kind of income redistribution on its own.

  • Government Solution: The government can address this by taxing its citizens to fund public redistribution programs.

  • Social Assistance: The most recognised method for redistributing income to low-income citizens.

    • Definition: Government programs designed to provide cash payments to vulnerable, low-income populations.

Social Assistance Programmes in Detail

  • In-kind programmes: Deliver specific goods, such as medical care or housing, directly to recipients.

  • Cash programmes: Provide direct cash benefits to recipients.

  • Benefit guarantee (G): The maximum cash welfare benefit available to individuals who have no other income.

  • Benefit reduction rate (t): The rate at which welfare benefits are reduced for each Rand of other income earned.

Social Development Spending in South Africa (2023 National Treasury Data)

  • Total Social Development Budget (2022/23 Revised Estimate): R378.5 billion.

    • Social protection expenditure: R264.363 billion (88.0% of which were social grants).

    • Social grants breakdown (2022/23):

      • Old-age grant: R92.067 billion (projected to grow to R114.013 billion by 2025/26, with an average annual growth of 7.4%).

      • Child-support grant: R77.179 billion (projected to grow to R93.034 billion by 2025/26, with an average annual growth of 6.4%).

      • Disability grant: R24.807 billion (projected to grow to R30.002 billion by 2025/26, with an average annual growth of 6.5%).

      • Foster care grant: R4.062 billion (projected to decrease to R3.362 billion by 2025/26, with an average annual decline of -6.1%).

      • Care dependency grant: R3.870 billion (projected to grow to R4.751 billion by 2025/26, with an average annual growth of 7.1%).

      • Grant-in-aid: R1.896 billion (projected to grow to R2.822 billion by 2025/26, with an average annual growth of 14.2%).

      • Social relief of distress: R29.104 billion (projected to decrease to R424 million by 2025/26, with an average annual decline).

    • Provincial social development: R21.257 billion.

    • Policy oversight and grant administration: R10.9 billion.

    • Social security funds (total): R93.424 billion.

      • Road Accident Fund: R50.322 billion.

      • Unemployment Insurance Fund: R32.801 billion.

      • Compensation funds: R10.301 billion.

Social Safety Net Spending (International Comparison - World Bank 2018)

  • South Africa's social safety net spending as a percentage of GDP is relatively high, ranking third in Sub-Saharan Africa and significantly above the regional mean (1.5%).

    • South Africa: Approximately 3.8% of GDP.

    • Lesotho: Approximately 4.2% of GDP.

    • Mauritius: Approximately 4.5% of GDP.

  • Global average for Social Safety Net spending (excluding health fee waivers) is approximately 1.5% of GDP.

Social Grant Beneficiary Numbers (South Africa - National Treasury 2023)

  • Total beneficiaries (2022/23): 26,604 thousand (projected to decline to 19,582 thousand by 2025/26, primarily due to the COVID-19 SRD grant expiring).

  • Breakdown by grant type (2022/23):

    • Child support: 13,283 thousand (approx. 62.0% of total).

    • Old age: 3,872 thousand (approx. 18.5% of total).

    • Disability: 1,051 thousand (approx. 4.9% of total).

    • Foster care: 267 thousand (approx. 1.0% of total).

    • Care dependency: 159 thousand (approx. 0.8% of total).

    • COVID-19 SRD grant: 7,971 thousand (approx. 12.8% of total, for 2022/23).

Distribution of Special COVID-19 Social Relief of Distress Grant (June 2020)

  • Analysis by Kohler and Bhorat (2020) indicated that the COVID-19 SRD grant was highly effective in targeting the poorest.

  • The highest proportion and absolute number of recipients were concentrated in the lowest income deciles.

    • The poorest 10% of per capita household income received the largest share, with an estimated absolute number of recipients exceeding 300,000.

    • The number of recipients steadily decreased as income deciles increased, showing strong progressivity.

Effects of Social Grants

Social Grants as a Mechanism to Increase Welfare for Poor Households

  • Social grants have become an increasingly vital source of income for poor South Africans.

  • This is especially critical given that poor households often lack access to sufficient wage income.

Social Grants and Incentives: The Work Disincentive

  • A key concern is the potential for perverse incentives, specifically the disincentive to work due to income transfers.

  • This relates to consumer utility maximisation and how transfers can alter an individual's budget constraint and labor-leisure choice.

The Labour Supply Model: The Budget Constraint

  • Example Scenario (Ayanda):

    • Wage (w): R10 per hour.

    • Total available hours in a year: 2,000 hours.

  • Budget Constraint (without transfers):

    • If Ayanda takes no leisure (works 2,000 hours), her maximum consumption is R20,000 per year (2,000 ext{ hours} imes R10/ ext{hour}). This is point A on the graph.

    • If she takes 2,000 hours of leisure (works 0 hours), her consumption is R0. This is point C on the graph.

    • The slope of this budget constraint is -10, representing the relative price of leisure in terms of consumption.

    • The x-axis can be interpreted as leisure hours (left to right) or hours of work (right to left).

Effects of a Means-Tested Transfer Programme on the Budget Constraint

  • Transfer Formula: For means-tested programs, the benefit ( B ) paid to an individual is calculated as: \$B = G - (t \times w \times h)

    • G: Benefit guarantee level.

    • t: Benefit reduction rate.

    • w: Wage rate.

    • h: Hours worked.

  • Example Scenario (Ayanda with transfers):

    • Benefit guarantee (G): R5,000.

    • Benefit reduction rate (t): 50%.

  • New Budget Constraint:

    • Even if Ayanda does not work (0 hours worked), she still receives R5,000 in consumption, establishing point D on the graph.

    • The segment of the budget constraint from D to B becomes flatter.

    • If Ayanda works an additional hour, she earns R10 in wages but loses R5 in benefits (50% of R10).

    • The effective slope in this segment becomes -5 (or -(R10 imes (1-0.5)) ), representing (1-t) \times w .

  • Eligibility Cutoff Point:

    • The budget constraint remains unchanged over the original segment (A to B) once Ayanda's earnings reach a certain threshold.

    • She is no longer eligible for the benefit once her earnings exceed R10,000, because \$R5,000 / 0.5 = R10,000.

    • To earn R10,000 at R10 per hour, she must work 1,000 hours (10,000 / 10).

    • This means she has 1,000 leisure hours (2,000 - 1,000).

Effects on Labour Supply

  • Scenario 1: Initially working more than 1,000 hours (e.g., at point E, working 1,600 hours / 400 leisure hours):

    • The transfer program has no effect on her labour supply, as her chosen work hours fall within the segment of the budget constraint that remains unchanged (AB).

    • She continues to work 1,600 hours and earns R16,000.

  • Scenario 2: Initially working less than 1,000 hours (e.g., at point F):

    • The transfer program reduces her labour supply due to two effects:

      • Income Effect: The transfer makes her effectively richer, leading her to consume more of all goods, including leisure. An increase in leisure means a decrease in hours worked.

      • Substitution Effect: The benefit reduction rate effectively lowers the opportunity cost of not working (the wage). The effective wage falls from R10 to R5. Since leisure is now relatively cheaper, she consumes more leisure, leading to a decrease in hours worked.

How Large Will the Labour Supply Response Be?

  • The magnitude of the labour supply response depends on:

    • The income earned by the individual before the benefit change.

    • The size of the income and substitution effects on their leisure/labour decision.

  • Individual Preferences (Indifference Curves):

    • The slope of the indifference curve, representing the Marginal Rate of Substitution (MRS) between leisure and consumption (MRS = MU{ ext{leisure}} / MU{ ext{consumption}}), is crucial.

    • Example (Peter vs. Ayanda): If Peter values leisure more highly relative to consumption than Ayanda, his indifference curve will be steeper.

    • Consequently, Peter will likely reduce his hours worked by more than Ayanda under the influence of the transfer program, even if starting from similar initial conditions.

Labour Supply Decisions with a 100% Benefit Reduction Rate

  • If the benefit reduction rate is 100% (t=1), any amount earned (w \times h) directly reduces the benefit (B) by the same amount, until the benefit is entirely phased out.

  • This creates a strong disincentive to work.

  • All individuals with income below the guarantee level, and many even with income above it, might immediately stop working to maximise leisure and consumption (by receiving the full guarantee without effort).

  • For example, with a guarantee G = R11,170 and a wage -12.50 (implicit wage rate), the budget constraint fundamentally changes, leading to different optimal points for leisure/work.

Direct Cash Transfers Are Strongly Progressive

  • In South Africa, 69% of all cash transfers are directed to the bottom 40% of the income distribution.

  • Grants, in decreasing order of progressivity, include:

    • Child support grant

    • Disability grant

    • Old age grant

  • All these grants demonstrate a high degree of progressivity.

  • Targeting Effectiveness: By directing grants to households with families and children, the system effectively targets the poor, as these demographics (households with elderly and school-aged children) are disproportionately represented at the lower end of the income distribution.

  • Consideration: This raises questions about how effectively other poor households, without children or elderly members, are targeted by the current grant scheme.