How might Matching Defined Contributions i help hl improve social protection design and labor market outcomes in MDCs? Ian Walker, The World Bank
Motivation and main messages 2
Adapting social protection to labor market realities In general, there is a need to rethink social il insurance programs and adapt them to labor market dynamics. Given the mobility of the labor force across occupations and between the formal and informal sectors, contributory systems that cover only formal sector workers are insufficient. However, having a parallel contributory and non contributory system does not seem to be the most efficient strategy either. Thus, policymakers need to come up with integrated systems that do not discriminate between workers depending on where they work (sector or occupation), while ensuring that the design encourages, rather than discourages, formal sector work. (RRW, p.83) 3
Policy question How can we expand the coverage of contributory tib t social insurance while minimizing labor market distortions and ensuring horizontal equity? Target group: informal workers, outside SI, on whom we are unable to impose savings mandates. Specific target group: informal workers with some savings capacity: the missing middle between formal sector and extreme poor beneficiaries of targeted anti poverty programs. 4
Main messages (1) Bismarkian systems are flawed: Low coverage, exclude informal sector and selfemployed. Rditibti Redistribution is non transparent tand often regressive. Tax element of contributions distorts labor markets, foments informality. Parallel non contributory systems help, but have drawbacks: Create a two tier system in terms of benefit levels. Create further incentives for informality. Fail to achieve complete coverage ( missing middle ). 5
Main messages (2) Ideal solution: a universal integrated system: Anybody can enroll. Contributions linked to benefits. Rditibti Redistribution is explicit it and targeted t dbased on needs. Financing is through contributions plus general taxation. But that might be hard to get, for myriad reasons Alternative : Extend reformed contributory arrangements into the informal sector. Thereby, limit the scope of the non contributory system. Matching defined contributions can be a key incentive. atc g de ed co t but o s ca be a ey ce t e. 6
Outline Typology of social insurance systems and the problem of coverage. Labor market distortions imposed by SI systems. A general policy framework to design and expand SI coverage. Potential role of MDCs 7
TYPOLOGY OF SI SYSTEMS AND THE PROBLEM OF COVERAGE 8
Mapping SI systems RISK PROGRAMS INSTRUMENTS S FINANCING MECHANISMS S Pensions Old age Disability Survivors Savings Risk pooling (insurance) Pay roll taxes Individual contributions Unemployment benefits Rdi Redistributionib i General revenues Health benefits Explicit Implicit
Systems that we observe Pensions Defined Benefit (risk pooling) (pay roll and IC) (implicit redistribution) Defined Contributions (savings) (pay roll and IC) (explicit redistribution) Unemployment benefits Severance Pay (risk pooling) (employer tax) (implicit redistribution) Unemployment Insurance (risk pooling) (pay roll and IC) (implicit redistribution) Unemployment Savings Accounts (savings) (pay roll and IC) (explicit redistribution) Health benefits Social H. Insurance (ik (risk pooling) (pay roll and IC) (implicit redistribution) National Health Services (ik (risk pooling) (general revenues) (implicit redistribution) Health Savings Accounts (savings) (pay roll and IC) (explicit redistribution)
Only a minority has access to SI Coverage (active members / labor force) 0 to 25 25 to 50 50 to 75 75 to 100 11
Informality is part of the problem Argentina Bolivia Brazil Chile Costa Rica Dominican Rep. Ecuador El Salvador Guatemala Mexico Nicaragua Panama Paraguay Peru Uruguay Venezuela SSE Formal salaried USE Informal salaried 0% 20% 40% 60% 80% 100% Share of Employment 12
Coverage is also correlated with income 100 80 60 40 20 0 Paraguay Nicaragua Honduras Dom. Rep. Bolivia Peru Ecuador Guatemala El Salvador 100 80 60 40 20 0 Colombia Venezuela Mexico Argentina Panama Brazil Uruguay Costa Rica Chile 13
Even those covered are not covered all the 14 time 100 Contribution Density (% of active life) 90 80 70 60 50 40 Q5 30 20 10 0 Argentina Chile Uruguay Source: Ribe, Robalino, and Walker (2010) Q1
15 Non contributory systems help but total coverage is still far from universal Pension Coverage by Income quintiles Bolivia 100 100 Chile 80 80 60 60 40 40 20 20 0 Q1 Q2 Q3 Q4 Q5 0 Q1 Q2 Q3 Q4 Q5 Costa Rica 100 80 60 40 Non contributory pensions Both contributory and noncontributory pensions 20 Contributory pensions 0 Q1 Q2 Q3 Q4 Q5 Source: Ribe, Robalino, and Walker (2010)
16 leaving a missing middle, in health as in old age income coverage, in most places 100 80 60 40 20 Health Insurance Coverage by Income Decile Colombia Peru 100 80 60 40 20 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 Chile 100 Non contributory Health Insurance 80 60 40 20 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 Both Contributory and Non contributory Health Insurance Contributory Health Insurance Source: Ribe, Robalino, and Walker (2010)
SI AND LABOR MARKET INCENTIVES 17
What are the possible distortions? ON THE SUPPLY SIDE Incentives to search for and keep jobs Unemployment Benefits Lack of access to unemployment benefits Incentives to switch jobs Lack of SI benefits portability Tax component of SS contributions Incentives for informal vs. formal work Subsidies to informal workers Pensions Incentives for participation in the labor force
What are the possible distortions? ON THE DEMAND SIDE Incentives to destroy jobs Severance pay Incentives to create jobs Pay roll tax Incentives for informal vs. formal sector
Informal Sector Productivity, Payroll Taxes, EMTRs, and Incentives for Informality or productiv vity in the in nformal Lab sect tor relative to the forma al sector B 1 EMTR 1 T 1 1 EMTR T B 1 B 2 EM TR2 EM TR 1 Individual prefers informal sector jobs Individual prefers formal sector jobs Slope = 1/t Social security contribution paid by employee Source: Authors calculations. Note: The slope of the line depends on the employer s payroll tax (T), which affects the total cost of labor in the formal sector, here assumed to be equal to labor productivity. The interception of the lines with the Y axis depends on the employee contribution rate, the value of the transfer (B), and the EMTR. Along the lines, the differences in labor productivity between the formal and informal sectors are equal to the tax wedge. Along the dotted d line, the EMTR is higher h than along the solid line. 20
Tax wedge is high in many countries Egypt Colombia Mexico Hungary Djibouti Romania Lithuania Czech R. Poland Turkey Slovenia Latvia Serbia and Slovak R. Macedonia Estonia Croatia Venezuela Ukraine Bulgaria Armenia Uzbekistan Argentina Algeria Tunisia Peru Morocco Belarus Bosnia Albania Iran Moldova Uruguay Kyrgyz R. Brazil Russian Azerbaijan Tajikistan Costa Rica Kazakhsta Yemen Georgia Jordan Ecuador Paraguay Libya Lebanon Niger Chile Iraq Bahrain World Average World Average 0 10 20 30 40 50 60 Tax Wedge 21
Severance pay can also be costly Colombia Peru Mexico Ecuador Uruguay Guatemala El Salvador Bolivia Argentina Panama Chile Honduras Venezuela Nicaragua Paraguay 0.0% 1.0% 2.0% 3.0% 4.0% 5.0% 6.0% 7.0% 8.0% 9.0% Insurance Equivalent Cost of Severance Pay (% wage bill) 22
CONCLUSIONS 23
Expanding SI coverage Expanding SI coverage requires opening access beyond the formal sector where membership can be imposed All citizens/residents would register in the same social insurance system regardless of where they work. SI systems should generally link contributions to benefits Defined contributions can reduce tax element, reduce gaming and help increase voluntary demand for SI This does NOT require funding, but it DOES require transparency, credibility dbl and generous rates of return 24
The role of matching contributions Impossible to enforce enrollment outside formal sector Incentives needed for those with low productivity, limited savings capacity and high personal discount rates Subsidies should ideally be: Explicit, financed as much as possible from general revenues Targeted (h (cheaper) with gradual withdrawal rates (minimizing EMTR; distortion is then contained and better managed) Ex ante (to incentivize contributions) rather than ex post (potentially reducing savings demand) Matching contributions can produce attractive returns in the very short term and help offset skepticism 25
What else might be needed to make this work on the ground? Flexibility in required contributions, to respond to fluctuating incomes Aggregators to lower transaction costs Promotion and marketing strategies Mobile agencies,, to reach out to the potential market Institutional credibility 26
Thank you 27