Defined Contribution Plans, Defined Contribution Plans, Mortality Risk, and the Demand Mortality Risk, and the Demand

for Annuities for Annuities

James PoterbaJames PoterbaMIT and NBERMIT and NBER

Longevity 5 Conference, September 2009 Longevity 5 Conference, September 2009

OutlineOutline

Shifting Composition of Retirement Shifting Composition of Retirement Saving: Rise of Defined Contribution Plans Saving: Rise of Defined Contribution Plans

Current Patterns of Payouts and Annuity Current Patterns of Payouts and Annuity Demand in Self-Directed DC Plans Demand in Self-Directed DC Plans

Implications of DC Plan Growth for the Implications of DC Plan Growth for the Demand for Longevity Insurance Demand for Longevity Insurance

U.S. Private Retirement System: U.S. Private Retirement System: Shift from DB to DC Shift from DB to DC

1980: Roughly 75% of Pension Contributions in 1980: Roughly 75% of Pension Contributions in the U.S. to Defined Benefit Plansthe U.S. to Defined Benefit Plans

2005: 73% of Pension Contributions to Defined 2005: 73% of Pension Contributions to Defined Contribution (401(k), 403(b)) Style PlansContribution (401(k), 403(b)) Style Plans

DC Plan and IRA Assets in 2006: $8.3T ($16.4T DC Plan and IRA Assets in 2006: $8.3T ($16.4T in Total Retirement Assets)in Total Retirement Assets)

Future Retirees will Have Lifetime Exposure to Future Retirees will Have Lifetime Exposure to 401(k)s401(k)s

Retirement Asset Landscape, 2006 Retirement Asset Landscape, 2006

401(k) Plans: $2.7T401(k) Plans: $2.7T

457 and 403(b) Plans: $834B 457 and 403(b) Plans: $834B

IRAs: $4.3TIRAs: $4.3T

Private DB Plans: $2.3TPrivate DB Plans: $2.3T

State/Local DB Plans: $3.0TState/Local DB Plans: $3.0T

Federal Plans: $1.1TFederal Plans: $1.1T

Private-Sector Pension Private-Sector Pension Participation Rates, 2008 Participation Rates, 2008

Full-Time Workers:Full-Time Workers:– 24% in Defined Benefit Plans24% in Defined Benefit Plans– 51% in Defined Contribution Plans51% in Defined Contribution Plans– 60% in Some Pension Plan60% in Some Pension Plan

Part-Time Workers: Part-Time Workers: – 10% in Defined Benefit Plans10% in Defined Benefit Plans– 18% in Defined Contribution Plans18% in Defined Contribution Plans– 23% in Some Pension Plan 23% in Some Pension Plan

401(k) Eligibility & Participation 401(k) Eligibility & Participation Rates, 1984-2003 SIPPRates, 1984-2003 SIPP

0

10

20

30

40

50

60

70

1984 1987 1991 1993 1995 1998 2003

401(k) Participation

401(k) Eligibility

Cohort Patterns of 401(k) Cohort Patterns of 401(k) Participation Participation

0

10

20

30

40

50

60

70

25 30 35 40 45 50 55 60

Age

Part

icip

ati

on

Rate

C15

C20

C25

C30

C35

C40

C45

C50

C55

C60

401(k) Balances Among 401(k) 401(k) Balances Among 401(k) ParticipantsParticipantsAverageAverage MedianMedian

19991999 $66,660$66,660 $24,844$24,844

20032003 80,59280,592 43,12743,127

20072007 137,430137,430 76,94676,946

Accumulation Potential in 401(k)s Accumulation Potential in 401(k)s (Purcell 2007)(Purcell 2007)

Married Couple Contributing 8% of Salary Married Couple Contributing 8% of Salary for 30 Yearsfor 30 Years

Historical Returns (pre-2008) on EquitiesHistorical Returns (pre-2008) on Equities

Median Balance at 65 for Median Earner: Median Balance at 65 for Median Earner: $468,000$468,000

2525thth Percentile Earner: $289,000; 75 Percentile Earner: $289,000; 75thth Percentile $706,000 Percentile $706,000

Median Replacement Rates for Median Replacement Rates for 401(k)-Participating Retirees, 2035 401(k)-Participating Retirees, 2035

(Holden & VanDerhei, 2005)(Holden & VanDerhei, 2005)

Income QuartileIncome Quartile Social SecuritySocial Security 401(k) Plan401(k) Plan

LowestLowest 52%52% 51%51%

SecondSecond 3131 5454

ThirdThird 2323 5959

HighestHighest 1616 6767

Ratio of Projected Defined Ratio of Projected Defined Contribution Plan Assets to Defined Contribution Plan Assets to Defined

Benefit Plan Liabilities Benefit Plan Liabilities

0

0.5

1

1.5

2

2.5

3

3.5

4

2000 2010 2020 2030 2040

Recent Events and DC Recent Events and DC Accumulation Profile Accumulation Profile

401(k)s Lost $500B Between June 30 and 401(k)s Lost $500B Between June 30 and December 31, 2008December 31, 2008Employer Cutbacks in 401(k) GenerosityEmployer Cutbacks in 401(k) GenerosityRegulatory Swing Away from DC Model – But to Regulatory Swing Away from DC Model – But to What? What? Open Issues: Incentives for Expansion of Open Issues: Incentives for Expansion of Pension Coverage, “Auto-Enrollment,” Pension Coverage, “Auto-Enrollment,” Government-Funded Plans for Low Income Government-Funded Plans for Low Income Households Households

The New Focus: Drawing Down The New Focus: Drawing Down Assets in Retirement Assets in Retirement

Search for Simple Rules: X% Per Year? Search for Simple Rules: X% Per Year? Simple Rules May Be Dominated Simple Rules May Be Dominated

With Uncertain Longevity, No Bequest With Uncertain Longevity, No Bequest Motives or Medical Spending Shocks, Motives or Medical Spending Shocks, Individuals Should Fully Annuitize Individuals Should Fully Annuitize

More Complex Analysis: Potential Medical More Complex Analysis: Potential Medical and Nursing Home Costs, Bequest and Nursing Home Costs, Bequest Motives – Annuity Demand is ReducedMotives – Annuity Demand is Reduced

Longevity Risk is a Key Factor in Longevity Risk is a Key Factor in Translating Assets into Lifetime Translating Assets into Lifetime

Income Security Income Security Conditional on Age 65, Probability of Dying by 75 is 0.254 for Men, 0.189 for Women

Conditional on Age 65, Probability of Living to 90 is 0.181 for Men, 0.275 for Women

Dispersion of Longevity Outcomes Dispersion of Longevity Outcomes for Married 65-Year-Old Couple for Married 65-Year-Old Couple 38.4 Expected Remaining Person-Life-Years

13.5 Expected Years Together

Expected Years Lived by Widowed Wife: 6.8

Expected Years Lived by Widower: 4.5

First Death: 25% Chance by Age 73, 50% by Age 78

Second Death: 50% Chance After Age 90, 25% After Age 93

Prob(Wife Survives Husband) = 57.5%

Factors the Influence Annuity Factors the Influence Annuity Demand Demand

Understanding of Mortality RiskUnderstanding of Mortality Risk

Employer Provision of Annuity Options to Employer Provision of Annuity Options to Employees in DC PlansEmployees in DC Plans

The Cost of AnnuitiesThe Cost of Annuities

Demand for Late-Life Liquid AssetsDemand for Late-Life Liquid Assets– Bequest MotivesBequest Motives– Concerns Related to Medical CostsConcerns Related to Medical Costs

Mortality Perceptions Mortality Perceptions

Hurd / McGarry Compare Subjective Hurd / McGarry Compare Subjective Mortality Probabilities in Health and Mortality Probabilities in Health and Retirement Survey with Actual Mortality Retirement Survey with Actual Mortality RatesRates

Men: Survey Average Survival Probability Men: Survey Average Survival Probability to Age 75: 0.622. “Actual” from Mortality to Age 75: 0.622. “Actual” from Mortality Table: 0.594. Women: 0.663 and 0.746.Table: 0.594. Women: 0.663 and 0.746.

Survival to 85: Subjective 0.388 for Men Survival to 85: Subjective 0.388 for Men (0.242 “actual”), 0.460 (0.438) for Women (0.242 “actual”), 0.460 (0.438) for Women

Life Expectancy for 65-year-old Life Expectancy for 65-year-old Males, by Birth Cohort Males, by Birth Cohort

0

5

10

15

20

25

1912 1917 1922 1927 1932 1937 1941

Top 50%

Bottom 50%

What Do DC Plan Participants What Do DC Plan Participants Choose? Choose?

DB Plan Evidence (Mottola/Utkus): Active DB Plan Evidence (Mottola/Utkus): Active De-Annuitization and Lump Sum Demand De-Annuitization and Lump Sum Demand ICI Study 1995-2000 Period: Only 23% of ICI Study 1995-2000 Period: Only 23% of DC Plan Participants with Choice at DC Plan Participants with Choice at Retirement Chose AnnuityRetirement Chose AnnuityOregon Public Employees Study Oregon Public Employees Study (Chalmers/Reuter) – Poor Health, Strong (Chalmers/Reuter) – Poor Health, Strong Recent Stock Market Returns Reduce Recent Stock Market Returns Reduce Annuity DemandAnnuity Demand

Annuity Choices of TIAA-CREF Annuity Choices of TIAA-CREF Participants, 1989-2001Participants, 1989-2001

0

10

20

30

40

50

60

1989 1991 1993 1995 1997 1999 2001

1-Life Annuity

Joint Annuity

Non-Annuity

Payout Options in Large Defined Payout Options in Large Defined Contribution Plans, 2000 NCS Contribution Plans, 2000 NCS

Only 38% of 401(k) Plans, 33% of All Only 38% of 401(k) Plans, 33% of All Defined Contribution Plans Defined Contribution Plans OfferOffer an an Annuity Option Annuity Option

Lump-Sum Distribution is the ONLY Lump-Sum Distribution is the ONLY Option in 28% of 401(k) Plans, 30% of All Option in 28% of 401(k) Plans, 30% of All DC PlansDC Plans

Why?Why?

Offering Annuities Imposes Offering Annuities Imposes Fiduciary Liabilities on Employers Fiduciary Liabilities on Employers

Annuitizing Annuitizing withinwithin DC Plan: Employer Must DC Plan: Employer Must Select Annuity ProviderSelect Annuity Provider

Long-Term Financial Health Difficult to Long-Term Financial Health Difficult to Gauge, Risk of Subsequent Lawsuit by Gauge, Risk of Subsequent Lawsuit by EmployeesEmployees

Annuity Options Available to Annuity Options Available to Individuals Individuals

Sales of New Single-Premium Immediate Sales of New Single-Premium Immediate Annuities: $12.8 Billion in 2007 Annuities: $12.8 Billion in 2007

Variable Annuity Market is Much Larger Variable Annuity Market is Much Larger but Very Few Assets are Annuitized but Very Few Assets are Annuitized

2007 SCF: 9.5% of Households Aged 75-2007 SCF: 9.5% of Households Aged 75-84 Report Receiving Income from 84 Report Receiving Income from AnnuitiesAnnuities

Annuitized Income Streams Approximately Annuitized Income Streams Approximately Half of Income for 65-74 Year Olds Half of Income for 65-74 Year Olds

Explaining Small Private Annuity Explaining Small Private Annuity Markets Markets

““DEMAND:” Precautionary Demand for DEMAND:” Precautionary Demand for Liquid Wealth, Bequest Motives, Informal Liquid Wealth, Bequest Motives, Informal Longevity Insurance Provided within Longevity Insurance Provided within FamiliesFamilies

““SUPPLY:” Unattractive Annuity Prices SUPPLY:” Unattractive Annuity Prices Because of Adverse Selection or Limited Because of Adverse Selection or Limited CompetitionCompetition

Expected Present Discounted Expected Present Discounted Value (EPDV) of Annuity Payouts Value (EPDV) of Annuity Payouts

per Premium Dollar per Premium Dollar

EPDVEPDVNOMNOM = = ΣΣt=1,Tt=1,TSStt*A*ANOMNOM/{/{ΠΠj=1,tj=1,t(1+i(1+ijj)})}

Survival Rates – Population Life Table or Survival Rates – Population Life Table or Annuitant Life Table, Projected Forward Annuitant Life Table, Projected Forward

Choice of Bonds for Discount Rates – Choice of Bonds for Discount Rates – Riskless Treasuries vs. Risky Corporates Riskless Treasuries vs. Risky Corporates

EPDV Does Not Recognize Insurance EPDV Does Not Recognize Insurance Value of Annuity to IndividualValue of Annuity to Individual

Money’s Worth of Individual Money’s Worth of Individual Annuities, May 2009 Annuities, May 2009

Annuitant Annuitant Mortality TableMortality Table

Population Population Mortality TableMortality Table

Interest Interest Rate:Rate:

Cor-Cor-porateporate

TreasuryTreasury Cor-Cor-porateporate

TreasuryTreasury

Men Age 65Men Age 65 0.8180.818 1.0541.054 0.7510.751 0.9460.946

Women Age Women Age 6565

0.8330.833 1.1031.103 0.7480.748 0.9600.960

Why Are EPDV Values < 1? Why Are EPDV Values < 1?

Insurance Company Administrative Costs Insurance Company Administrative Costs or Profitsor Profits

Adverse Selection: Annuitant Population is Adverse Selection: Annuitant Population is Longer-Lived Than Population at LargeLonger-Lived Than Population at Large

Risk Premium to Cover Cost of Future Risk Premium to Cover Cost of Future Mortality Improvement Mortality Improvement

Adverse Selection: Comparing Adverse Selection: Comparing Mortality Rates for Annuitants & Mortality Rates for Annuitants &

Population at Large, 2007 Population at Large, 2007

Annuitant MortalityAnnuitant Mortality Population MortalityPopulation Mortality

MenMen WomenWomen MenMen WomenWomen

6565 1.02%1.02% 0.57%0.57% 1.72%1.72% 1.16%1.16%

7575 2.982.98 1.611.61 4.294.29 2.982.98

8585 8.068.06 5.085.08 11.3511.35 8.548.54

Evidence of Adverse Selection: UK Evidence of Adverse Selection: UK Compulsory Annuity Market Compulsory Annuity Market

Nominal AnnuityNominal Annuity

Inflation-Indexed Annuity Inflation-Indexed Annuity

““Escalating” Annuity (3% per year)Escalating” Annuity (3% per year)

Nominal Annuity: For 65 year old males, Nominal Annuity: For 65 year old males, 41% of EPDV is in First Five Years, 6% 41% of EPDV is in First Five Years, 6% Beyond Age 85; Contrast with 34% (9%) Beyond Age 85; Contrast with 34% (9%) for a Policy with 3%/Year Escalation (US for a Policy with 3%/Year Escalation (US 2008 Corporate Discounting)2008 Corporate Discounting)

Five-Year Survival Probability, 61-Five-Year Survival Probability, 61-65 Year Old Male Annuitant 65 Year Old Male Annuitant

(Finkelstein-Poterba)(Finkelstein-Poterba)Compulsory Compulsory AnnuityAnnuity

Voluntary Voluntary AnnuityAnnuity

Nominal Nominal 0.9130.913 0.9510.951

EscalatingEscalating 0.9700.970 0.9890.989

GuaranteedGuaranteed 0.9110.911 0.9400.940

Index-LinkedIndex-Linked 0.9620.962 0.9800.980

A Key Concern: Late Life Medical A Key Concern: Late Life Medical CostsCosts

CBO Projection for Cohort Turning 65 in CBO Projection for Cohort Turning 65 in 20102010

Any Nursing Home Use: 45%Any Nursing Home Use: 45%

One Year or Longer in a Nursing Home: One Year or Longer in a Nursing Home: 25%25%

Average Nursing Home Costs: $187/day Average Nursing Home Costs: $187/day for Semi-Private Room, $209/day for for Semi-Private Room, $209/day for Private Room Private Room

How Large is the Risk Premium for How Large is the Risk Premium for Aggregate Mortality Risk? Aggregate Mortality Risk?

Forecasting Mortality is Difficult Forecasting Mortality is Difficult

Risk of Medical Breakthrough Could Risk of Medical Breakthrough Could Change ExperienceChange Experience

Life Insurers are Affected by Illness Life Insurers are Affected by Illness Shocks (1918 Influenza, AIDS) Shocks (1918 Influenza, AIDS)

How Does Correlation with Other How Does Correlation with Other Aggregate Shocks Affect Risk Charges?Aggregate Shocks Affect Risk Charges?

Share of Annuity EPDV From Share of Annuity EPDV From Payouts in First Five Years (2009) Payouts in First Five Years (2009)

AnnuitantAnnuitant PopulationPopulation

Interest Interest RateRate

CorporateCorporate TreasuryTreasury CorporateCorporate TreasuryTreasury

Men, 65Men, 65 0.4020.402 0.3350.335 0.4310.431 0.3670.367

Women, Women, 6565

0.3720.372 0.3010.301 0.4070.407 0.3410.341

Annual Mortality Improvement RateAnnual Mortality Improvement Rate

1950-1950-19801980

1980-1980-20072007

1950-1950-20072007

Men, 65-74Men, 65-74 -0.53%-0.53% -1.61%-1.61% -1.05%-1.05%

Men, 75-84 Men, 75-84 -0.58-0.58 -0.99-0.99 -0.78-0.78

Men, 85+Men, 85+ -0.56-0.56 0.290.29 -0.14-0.14

Women, 65-74Women, 65-74 -1.61-1.61 -0.49-0.49 -1.07-1.07

Women, 75-84Women, 75-84 -1.57-1.57 -0.35-0.35 -0.98-0.98

Women, 85+ Women, 85+ -1.09-1.09 0.290.29 -0.42-0.42

Lee-Carter Mortality Model and Lee-Carter Mortality Model and Monte Carlo Results on EPDVs Monte Carlo Results on EPDVs

ln mln ma,ta,t = = ααaa + + aa*k*ktt + + εεa,t a,t with mwith ma,ta,t = crude death = crude death

rate at age a in year t rate at age a in year t

qqa,ta,t = mortality rate, age a, year t = 1 – exp[- m = mortality rate, age a, year t = 1 – exp[- ma,ta,t]]

Aggregate Mortality Factor kAggregate Mortality Factor k tt Follows a Random Follows a Random

Walk with DriftWalk with Drift

Estimate for {Estimate for {ααaa, , a , a , kktt} for Men over 1950-2007 } for Men over 1950-2007

Period (a = 65, …, 110) Period (a = 65, …, 110)

Stochastic Simulation of Future Paths of Stochastic Simulation of Future Paths of Mortality Rates Can be Used to Compute EPDV Mortality Rates Can be Used to Compute EPDV of Annuitiesof Annuities

Potential Variation in EPDV of Potential Variation in EPDV of Annuity for 65-Year-Old Men, 2009 Annuity for 65-Year-Old Men, 2009

Valuation Valuation PercentilePercentile

Treasury Yield Treasury Yield Discount Rate Discount Rate

Corporate Yield Corporate Yield Discount RateDiscount Rate

1010thth 0.8920.892 0.7170.717

2525thth 0.9140.914 0.7310.731

MedianMedian 0.9380.938 0.7450.745

7575thth 0.9610.961 0.7610.761

9090thth 0.9820.982 0.7730.773

Implications of Continued Growth of Implications of Continued Growth of Self-Directed DC Plans Self-Directed DC Plans

Absent Compulsory Annuitization (UK Absent Compulsory Annuitization (UK Model), Individuals Will Make Decisiosn Model), Individuals Will Make Decisiosn About Annuitization About Annuitization

Marketing Longevity Insurance Products to Marketing Longevity Insurance Products to Institutions Alone (Plan Managers, Institutions Alone (Plan Managers, Pension Fiduciaries) Will Not Penetrate Pension Fiduciaries) Will Not Penetrate DC UniverseDC Universe

Longevity Risk Facing Insurers is Only Longevity Risk Facing Insurers is Only One Factor Depressing Annuity Volume One Factor Depressing Annuity Volume