Nonparametric Evidence on the Effects of Financial Incentives on - PowerPoint PPT Presentation

Nonparametric Evidence on the Effects of Financial Incentives on Retirement Decisions Day Manoli Andrea Weber UT Austin & NBER University of Mannheim June 2013 Manoli and Weber () Financial Incentives at Retirement June 2013 1 / 37

Introduction How do individuals adjust labor supply in response to wage variation? Answer important in many fields: macro, public finance, labor ◮ Design of equilibrium models of the labor market ◮ Normative tax policy analysis ◮ Modeling labor supply ◮ Dispute over magnitude of elasticities between micro and macro studies Recent literature has highlighted the importance of distinguishing between the intensive and extensive margin (Heckman 1993, Saez 2002, Chetty et al. 2012) We focus on retirement decision in response to anticipated benefits Contribution: quasi-experimental estimates of extensive margin labor supply elasticities Manoli and Weber () Financial Incentives at Retirement June 2013 2 / 37

Introduction How do individuals adjust retirement entry to anticipated benefits? ◮ Relevant for design of pension systems, reforms, explaining retirement patters based on financial incentives ◮ Complicated to disentangle incentives from various policies at retirement Research design based on a simple and salient incentive structure ’independent’ from public pension system Exploit discontinuities in financial incentives along a dimension other than age Reduced form concept of extensive margin intertemporal substitution elasticities Noparametric estimation method based on bunching estimators (Saez 2009, Chetty et. al 2012, Kleven and Waseem 2013) Manoli and Weber () Financial Incentives at Retirement June 2013 3 / 37

Research Design Retirement rule in Austria: individuals who complete 10 years of tenure by retirement qualify for lump-sum payment from employer ◮ Simple and salient rule ◮ Benefits are fully anticipated but small relative to lifetime wealth ◮ Focus on delay in retirement entry decisions Examine retirement behavior around the thresholds ◮ Present graphical evidence on responses in retirement entry ◮ Examine heterogeneity across population groups Examine magnitude of financial incentive ◮ Contrast legislative incentive with estimated incentives from the data Elasticity relates retirement responses to financial incentives Manoli and Weber () Financial Incentives at Retirement June 2013 4 / 37

Institutional Background I Employer-provided retirement benefits within the severance pay system amount based on tenure at retirement mandated, lump-sum payments at retirement payments based on salary, not income / total compensation funds set aside by employers based on size of work force retirement is only voluntary separation that leads to payments severance payment is taxed at a constant, low rate of 6% Manoli and Weber () Financial Incentives at Retirement June 2013 5 / 37

Severance Payments as Fraction of Annual Salary 1 .75 .5 .333 0 0 5 10 15 20 25 30 Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 6 / 37

Institutional Background II Government-provided pensions: Normal (statutory) retirement ages: 65 (men) & 60 (women) Early retirement ages: 60 (men) & 55 (women) Ages 55-59: retirement through disability pensions Replacement rates ≈ 75% Even with bonuses for retirement at older ages, system is still actuarially unfair for most individuals Manoli and Weber () Financial Incentives at Retirement June 2013 7 / 37

Exits from the Labor Force into Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 8 / 37

Data Austrian Social Security Database 1972 - 2006 ◮ matched employer-employee census of private sector ◮ complete earnings and employment histories ◮ some demographic information on workers and firms Income Tax Records 1994 - 2005 ◮ employer reports to tax office at the end of the year ◮ annual salary plus withholdings of social security contributions and income taxes ◮ separate category for severance payments Sample restrictions: ◮ non-construction workers ◮ individuals still working at age 54 ◮ retiring within 6 months of last job ◮ with uncensored job-tenure at retirement ◮ retirements 1997-2005 matched to tax records Sample: 89,426 individual retirements with 6-28 years of tenure Manoli and Weber () Financial Incentives at Retirement June 2013 9 / 37

Summary Statistics # Individuals 89,426 Mean Std.dev Female 0.51 Retirement Age 59.1 2.59 Tenure 15.66 6.10 Annual Earnings 29,327 12,949 Severance Pay 18,510 21,661 Implicit Tax Rate 0.81 0.21 Years of Employment 34.3 8.67 Years of Sick Leave 0.17 0.33 Fractions: Disability Pension 0.21 Early Retirement 0.57 Old Age Pensions 0.22 Manoli and Weber () Financial Incentives at Retirement June 2013 10 / 37

Outline Empirical evidence on retirements Empirical evidence on severance payments Reduced form elasticity concept Estimation strategy Estimation results Discussion: interpretation and policy relevance Manoli and Weber () Financial Incentives at Retirement June 2013 11 / 37

Distribution of Tenure on Retirement 1000 800 Individuals 600 400 200 10 15 20 25 Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 12 / 37

Adjusting for Coviariates .12 .1 .08 .06 .04 .02 0 10 15 20 25 Years of Tenure Manoli and Weber () Financial Incentives at Retirement June 2013 13 / 37

Distribution of Job Starts by Age 12000 # of Individuals Starting New Jobs 10000 8000 6000 4000 2000 40 45 50 55 Age Women Men Manoli and Weber () Financial Incentives at Retirement June 2013 14 / 37

Hetergogeneity by Health Status Manoli and Weber () Financial Incentives at Retirement June 2013 15 / 37

Hetergogeneity by Age and Gender Male, Retire Age < 60 Male, Retire Age = 60 Male, Retire Age > 60 250 300 400 200 200 300 150 100 200 100 Individuals 100 50 0 Female, Retire Age < 60 Female, Retire Age = 60 Female, Retire Age > 60 800 150 300 600 100 200 400 100 50 200 0 0 0 10 15 20 25 10 15 20 25 10 15 20 25 Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 16 / 37

Distribution of Severance Payments, 10 Year Threshold A1. Tenure 9 years A2. Tenure 10 years 20 20 15 15 Percent Percent 10 10 5 5 0 0 0 .25 .33 .5 .75 1 0 .25 .33 .5 .75 1 Manoli and Weber () Financial Incentives at Retirement June 2013 17 / 37

Distribution of Severance Payments, 15 Year Threshold B1. Tenure 14 years B2. Tenure 15 years 20 20 15 15 Percent Percent 10 10 5 5 0 0 0 .25 .33 .5 .75 1 0 .25 .33 .5 .75 1 Manoli and Weber () Financial Incentives at Retirement June 2013 18 / 37

Distribution of Severance Payments, 20 Year Threshold C1. Tenure 19 years C2. Tenure 20 years 15 15 10 10 Percent Percent 5 5 0 0 0 .25 .33 .5 .75 1 0 .25 .33 .5 .75 1 Manoli and Weber () Financial Incentives at Retirement June 2013 19 / 37

Distribution of Severance Payments, 25 Year Threshold D1. Tenure 24 years D2. Tenure 25 years 15 15 10 10 Percent Percent 5 5 0 0 0 .25 .33 .5 .75 1 0 .25 .33 .5 .75 1 Manoli and Weber () Financial Incentives at Retirement June 2013 20 / 37

Distribution of Severance Payments A. 15th Percentile B. 25th Percentile 1 1 .8 .8 Sev Pay Fraction Sev Pay Fraction .6 .6 .4 .4 .2 .2 0 0 5 10 15 20 25 30 5 10 15 20 25 30 Years of Tenure at Retirement Years of Tenure at Retirement C. 50th Percentile D. 75th Percentile 1 1 .8 .8 Sev Pay Fraction Sev Pay Fraction .6 .6 .4 .4 .2 .2 0 0 5 10 15 20 25 30 5 10 15 20 25 30 Years of Tenure at Retirement Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 21 / 37

Definition of Responder Sample, 10 Year Threshold A1. 10 Year Threshold, Responders A2. 10 Year Threshold, Non-Responders 800 800 600 600 Individuals Individuals 400 400 200 200 0 0 7 8 9 10 11 12 13 7 8 9 10 11 12 13 Years of Tenure at Retirement Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 22 / 37

Definition of Responder Sample, 15 Year Threshold B1. 15 Year Threshold, Responders B2. 15 Year Threshold, Non-Responders 800 800 600 600 Individuals Individuals 400 400 200 200 0 0 12 13 14 15 16 17 18 12 13 14 15 16 17 18 Years of Tenure at Retirement Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 23 / 37

Definition of Responder Sample, 20 Year Threshold C1. 20 Year Threshold, Responders C2. 20 Year Threshold, Non-Responders 600 600 400 400 Individuals Individuals 200 200 0 0 17 18 19 20 21 22 23 17 18 19 20 21 22 23 Years of Tenure at Retirement Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 24 / 37

Definition of Responder Sample, 25 Year Threshold D1. 25 Year Threshold, Responders D2. 25 Year Threshold, Non-Responders 600 600 400 400 Individuals Individuals 200 200 0 0 22 23 24 25 26 27 28 22 23 24 25 26 27 28 Years of Tenure at Retirement Years of Tenure at Retirement Manoli and Weber () Financial Incentives at Retirement June 2013 25 / 37