capital (income) taxation and pension system reform
0 likes86 views
The document discusses the implications of pension privatization and capital income taxation, highlighting a gradual shift from defined benefit to defined contribution systems. It emphasizes that while privatization can enhance welfare in the long run, it may reduce insurance mechanisms within the pension system, and that capital income taxation can be beneficial for funding such reforms. The analysis indicates that optimal fiscal adjustments and understanding household responses to taxation are crucial for improving efficiency and achieving political support for pension reforms.
More Related Content
Similar to capital (income) taxation and pension system reform (20)
capital (income) taxation and pension system reform
- 1. Efficiency versus insurance: capital income taxation and privatizing social security Oliwia Komada (WSE and GRAPE) Krzysztof Makarski (NBP, WSE and GRAPE) Joanna Tyrowicz (GRAPE, IOS, UW, and IZA) European Economic Association – 2020 1 / 33
- 2. Motivation
- 3. Motivation – pension systems Problem: Longevity hazards pension system viability: for US deficit ↑ 1.4% of GDP Feldstein, BEA, SSA Solution: Privatization of pension system (redistributive) Defined Benefit ⇒ Defined Contribution • within generations redistribution −→ insurance ↓ • aggregate change in the economy −→ efficiency ↑ Consensus: • Deterministic setup: efficiency wins over the looses • Stochastic setup: insurance drags efficiency boost down Nishiyama & Smetters (2007, QJE) and subsequent literature • Reform ⇒ fiscal adjustments, elasticity of response to taxation is crucial for welfare Diamond and Spinnewijn 2011, Golosov et al. 2013, Straub and Werning 2020 2 / 33
- 4. Motivation – capital (income gains) taxation Why is taxing capital a good idea for pension privatization? Q1: Is it ok to raise τk? ←− in the short run Yes! Households have to save more, low elasticity of savings with respect to τk: • longevity ↑ ⇒ assets ↑ • redistribution ↓ ⇒ precautionary savings ↑ Q2: Is it ok to reduce τk? ←− in the long run Yes! We can do it! • system immune to longevity ⇒ tax ↓ in the long run Our contribution 1: Insights on τk in the context of longevity and pension privatization. 2: Quantification of the role of the insurance motive in terms of welfare. 3: Comparison of welfare and macro effects along with different fiscal tools. 3 / 33
- 5. What we do OLG model with idiosyncratic income shocks and longevity, US Baseline: US pension system = DB + redistribution Reform: gradual shift to (partially funded) DC • tax on capital income gains (with or without temporal debt adjustment) • compare it with the literature • consumption tax closure (with or without temporal debt adjustment) • 2 pension system adjustments: contributions or benefits • study welfare effect and political support • decompose welfare change into insurance and efficiency 4 / 33
- 6. Results in a nutshell 1. Privatization of social security + τk → welfare ↑: • in the short run: low elasticity of response to τk • in the long run: τk boosts efficiency more then other closures, e.g. τc • insurance motive actually rather small −→ Nishiyama & Smetters is NOT universal 2. Public debt often “buys” political support 3. Welfare gains and political support only sometimes overlap Results robust to • secular stagnation • more rapid reform implementation • different values of risk aversion 5 / 33
- 7. Model
- 8. Consumers • uncertain lifetimes: live for 16 periods, with survival rate πj,t < 1 • uninsurable productivity risk: + endogenous labor supply AR(1) process approximated by Markov chain • CRRA utility function • pay taxes (progressive on labor, linear on consumption and capital gains) • contribute to pensions, face natural borrowing constraint Firms and markets • Producers with Cobb-Douglas production function, with capital depreciation rate d • no annuity, financial markets with (risk free) interest rate 6 / 33
- 9. Government • Finances government spending Gt, constant as a share of GDP, • Balances pension system: paying subsidyt, • Services debt: ∆Dt + rtDt = Dt − Dt−1 + rtDt, • Collects taxes on capital, consumption, labor (progressive given by Benabou form), Gt + subsidyt + ∆Dt + rtDt = τk,trtAt + τc,tCt + Taxl,t −→ Which taxes to adjust? The elasticity of response to taxation important! 7 / 33
- 10. Pension system Baseline scenario: AIME = PAYG DB with progressivity • redistribution within a cohort ⇒ insurance but high labor distortion • defined benefit → longevity ⇒ long run deficit Reform scenario: (partially funded) DC • individual defined contribution ⇒ less insurance but less labor distortion • self balancing ⇒ no long run deficit • if partially funded ⇒ short run deficit but since r > g higher pensions 8 / 33
- 11. Results
- 12. Major effects of the reform Calibration Pensions linked to contributions 1. reduced labor supply distortion (efficiency ↑) 2. income shocks carry over to retirement (insurance ↓ ) Intuition: fiscal policy can reinforce or attenuate these effects Eventually, taxes decline (relative to the baseline scenario of permanent pension system deficit) Intuition: long run effects are surely positive, but elasticity matter for the size of these effects 9 / 33
- 13. Pension system subsidy and taxes pension subsidy τk tax rate 10 / 33
- 14. Pension system subsidy and taxes pension subsidy τk tax rate 11 / 33
- 15. Pension system subsidy and taxes pension subsidy τk tax rate 12 / 33
- 16. Pension system subsidy and taxes pension subsidy τk tax rate 13 / 33
- 17. Pension system subsidy and taxes pension subsidy τk tax rate 14 / 33
- 18. Pension system subsidy and taxes pension subsidy τktax rate 15 / 33
- 19. Why does τk work so nicely with longevity and funding? 16 / 33
- 20. Response of k to τk over transition - smaller in a short run 17 / 33
- 21. Response of k to τk over transition - bigger in the long run 18 / 33
- 22. Welfare effects of the pension system reform Reform scenario to DC to DC with partial funding Life expectancy constant rising constant rising Fiscal closure τk 1.25 1.37 -0.28 0.60 1. DC: • no fiscal cost • the only cost of the reform is insurance loss • drop in labor market distortion generates sufficient efficiency gain 2. DC with partial funding • insurance loss + fiscal cost >> efficiency gain due to ↓ labor distortion • welfare ↓ without longevity, with longevity ↑, additional efficiency gain due to lower capital tax in the long run • without rising longevity little or no merit in pension system privatization Replication of N& S results 19 / 33
- 23. Welfare across generations Reform: DC with partial funding τk τc 20 / 33
- 24. Welfare decomposition: insurance loss efficiency gain Reform: DC with partial funding τk τc 21 / 33
- 25. Debt smoothing helps with political support (not just for τk) 22 / 33
- 26. Debt smoothing helps with political support (not just for τk) 23 / 33
- 27. Debt smoothing helps with political support (not just for τk) 24 / 33
- 28. Sensitivity of the results 1. Capital income taxation vs. other fiscal closures details • in long run welfare ↑ for all closures • only τk + debt ⇒ aggregate welfare ↑ + political support 2. Declining rate of exogenous technological progress details • r >> g more long run benefits due to funding 3. Does it matter how fast the reform is implemented? details • so far: only future cohorts join the reformed P.S. • more rapid reform (45+ join new P.S.) ⇒ the key results the same 4. What is quantitative role for redistribution? details • high levels of risk aversion ⇒ insurance loss dominant, • the key argument: τk works better than τc is still valid 25 / 33
- 29. Conclusion
- 30. Summary 1. Fiscal cost + insurance loss efficiency gain if reform accompanied by complementary closure. 2. Distribution of fiscal cost and gains makes capital tax attractive closure. Why is capital tax so nice? −→ relatively less responsive to the tax hikes capital accumulation ↑ with longevity + funding + less insurance Without longevity: funding and DC make little sense Features of this literature • Savings have a roughly 10% reaction to longevity do people really “understand” the risk of old-age poverty • Labor has a roughly 10% reaction to reduced distortions do people really “understand” the link between contributions and pensions 26 / 33
- 31. Questions or suggestions? Thank you! w: grape.org.pl t: grape org f: grape.org e: [email protected] 27 / 33
- 32. Calibration to replicate US economy (2015) Preferences: instantaneous utility function take CRRA form with • Risk aversion in equal to 2 • Preference for leisure φ matches average hours 33% • Discounting rate δ matches interest rate 4.5% Idiosyncratic productivity shock based on Kruger and Ludwig (2013): • Persistence η = 0.95 • Variance ση = 0.375 Pension system • Replacement rate ρ matches benefits as % of GDP 5.2% • Contribution rate balances pension system in the initial steady state • Pension eligibility age at 65 (¯j = 9) Taxes {τc, τk, τl} match revenue as % of GDP {2.8%, 5.4%, 9.2%} Depreciation rate d matches investment rate of 22% 28 / 33
- 33. Welfare effects of the pension system reform Reform scenario to DC to DC with partial funding N&S Life expectancy constant rising constant rising constant constant rising (1) (1a) (2) (2a) (2b) (3) (3a) Fiscal closure τc 1.18 0.94 0.14 -0.06 -0.41 -0.52 -0.57 τk 1.25 1.37 -0.28 0.60 0.06 -0.24 0.09 1. DC: • no fiscal cost • the only cost of the reform is insurance loss • drop in labor market distortion generates sufficientefficiency gain 2. DC with partial funding • insurance loss + fiscal cost >>efficiency gain due to ↓ labor distortion • welfare ↓ without longevity, with longevity ↑, additionalefficiency gain due to lower capital tax in the long run 3. N&S • privatization = pension reduction, there is a fiscal cost of the reform • higher interest rate environment - future gains matter less 29 / 33
- 34. Capital income taxation vs. other fiscal closures Reform: DC + partial funding Fiscal closure Welfare effects Political support (in %) long run aggregate initial steady state capital income tax τk 1.93 0.60 42.60 capital income tax smoothed debt + τk 1.93 0.55 90.62 consumption tax τc 0.93 -0.06 35.16 consumption tax smoothed debt + τc 0.93 -0.07 74.20 raising contributions τ 1.21 -0.01 58.13 reducing pensions τb 0.07 0.11 9.85 • only capital income tax (with debt) and pension reduction yields to increase in welfare • but only capital income tax with debt gain political support 30 / 33
- 35. Secular stagnation In our main calibration, we assume a constant g = 2%. In robustness check, we assume a gradual decline in the rate of technological progress to 1.5%. Relatively high technological progress in our central scenario favors the current pension system in the US. With declining technological progress, its even easier to implement privatization and obtain welfare improvement and political support. Fiscal closure Welfare effects Political support final steady state aggregate initial steady s reducing pensions τb 0.12 0.11 9.85 raising contributions τ 1.76 0.05 58.13 capital income tax τk 2.05 0.39 42.60 capital income tax smoothed debt + τk 2.12 0.38 90.62 consumption tax τc 1.46 0.05 35.16 consumption tax smoothed debt + τc 1.46 0.06 74.21 31 / 33
- 36. Does it matter how fast the reform is implemented? In our main scenario, only newborns enter the privatized pension system. Faster implementation (workers older that 45 enter new pension) of the reform yields to higher pension deficit and, thus, the fiscal cost in the short run. Still, capital income taxation with debt smoothing generates welfare increase and political support. Fiscal closure Welfare effects Political support final steady state aggregate initial steady s reducing pensions τb 0.07 0.11 9.85 raising contributions τ 1.21 0.03 58.13 capital income tax τk 1.93 0.67 42.60 capital income tax smoothed debt + τk 1.93 0.55 90.62 consumption tax τc 0.93 0.00 42.60 consumption tax smoothed debt + τc 0.93 -0.07 74.21 32 / 33
- 37. What is a quantitative role for redistribution? • the main results: CRRA utility function with θ = 2 • with high levels of risk aversion (θ ≥ 3), the insurance channel becomes quantitatively dominant • but the key argument: τk works better than τc is still valid 33 / 33