wadepfau sequence of returns risk -...

9/10/2014

1

Sequence of Returns RiskSequence of Returns Risk

W d D Pf Ph D CFAWade D. Pfau, Ph.D., CFAThe American CollegeinStream SolutionsinStream SolutionsMcLean Asset Management

Retirement Researcher blog (wpfau.blogspot.com)

9/10/2014

2

Outline for Presentation

• Retirement Income Planning: Overview of new circumstances and risksDeep Dive into Sequence Risk

• Risk Management Techniques for Sequence Risk• Reduce Portfolio Volatility

1. Relationships between return, volatility, and spending2 Rising equity glidepaths during retirement2. Rising equity glidepaths during retirement

• Flexible Spending Strategies1. Decision Rules to Guide Retirement Withdrawals

9/10/2014

3

Retirement Risks

*

9/10/2014

4

Challenges for an I di id l P i PlIndividual Pension Plan

•Asset Returns•Asset Returns• Pension Manager – Pool returns across generations • Ad i O h k t th t• Advisor – One whack at the cat

Longevity Risk• Pension Manager – systemic increases in longevity• Advisor – Idiosyncratic longevity risk

9/10/2014

5

“[Retirement I Pl i ] iIncome Planning] is

a really hard problem. It’s the

hardest problem I’ve ever looked at.”

William SharpeWilliam SharpeCFA Institute Conference, 2014

9/10/2014

6

Sequenceofof

ReturnsRisk

Journal of Financial Service Professionals, January 2014

9/10/2014

7

Sequence Riskq

•Most vulnerable to returns when wealth is largest

•Pre retirement dollar cost•Pre-retirement dollar cost averaging reverses in distribution phase

9/10/2014

8

Sequence Risk - Accumulationq• Wealth Accumulation with Rolling 30-Year Savings Periods• Mean Return = 7%, Standard Deviation = 20%• With 15% i r t & t t l r E p t d W lth = 10• With 15% savings rate & constant salary, Expected Wealth = 10

9/10/2014

9

Sequence Risk - Distributionq• Maximum Spending with Rolling 30-Year Distribution Periods• Mean Return = 7%, Standard Deviation = 20%• With rk t p t ti E p t d Sp di r t = 6 2%• With market expectations, Expected Spending rate = 6.2%

9/10/2014

10

Withdrawal Rates: Explanatory P r f E rl d L t R t rPower of Early and Late Returns

25MWR = 4.34 + 0.41 * First10R2 77%

25MWR = 5.81 + 0.16 * Last20R2 5%

20

hdra

wal

Rat

e

R2 = 77%

20

hdra

wal

Rat

e

R2 = 5%

10

15

Sus

tain

able

With

10

15

Sus

tain

able

With

5

Max

imum

S

5

Max

imum

S

-20 -10 0 10 20 300

Compounded Return Over First 10 Years-20 -10 0 10 20 300

Compounded Return Over Final 20 Years

9/10/2014

11

Lifetime Sequence of Returns Riskq

14

16

n

10

12

14

h Y

ear's

Ret

urn

6

8

Pow

er fo

r Eac

h

2

4

Exp

lana

tory

0 10 20 30 40 50 600

Year

9/10/2014

12

L B d Yi ldLow Bond Yields=

Lower SustainableLower SustainableWithdrawal Rates

&

H i ht dHeightenedExposure to

Sequence Risk

Journal of Financial Planning, June 2013 (article link)

Sequence Risk

9/10/2014

13

Importance of current market conditions

• Sources of bond returns: • initial bond yield• subsequent yield changes

• Sources of stock returns: • dividend income • growth of the underlying earnings• changes in the valuation multiples placed on those

earningsearnings

9/10/2014

14

16

18

etur

ns

Figure 2Relationship Between Bond Yields and Subsequent Bond Returns

ITGB YieldSubsequent 5-Year Return 16

18

6

8

10

12

14

and

Sub

sequ

ent 5

-Yea

r Nom

inal

Re

6

8

10

12

14

sequ

ent 5

-Yea

r Nom

inal

Ret

urn

1920 1930 1940 1950 1960 1970 1980 1990 2000 20100

2

4

Bon

d Y

ield

s a

0 2 4 6 8 10 12 14

0

2

4

Sub

s

Intermediate-Term Government Bond Yield

Return = 0.3 + 1.07 * YieldR2 = 0.88

10

12

14

omin

al R

etur

ns

ITGB YieldSubsequent 10-Year Return

14

16

18

Ret

urn

4

6

8

Yie

lds

and

Sub

sequ

ent 1

0-Y

ear N

o

4

6

8

10

12

Sub

sequ

ent 1

0-Y

ear N

omin

al R

1920 1930 1940 1950 1960 1970 1980 1990 2000 20100

2

Bon

d Y

0 2 4 6 8 10 12 14

0

2

Intermediate-Term Government Bond Yield

Return = 0.75 + 0.96 * YieldR2 = 0.92

9/10/2014

15

Monte Carlo SimulationsFailure Rates for 4% Inflation-Adjusted Withdrawals over 30 Years

For Different Asset Return Assumptions30%

Stocks50%

Stocks70%

StocksHistorical Characteristics (means, volatilities, correlations) 1926- 6% 6% 8%Historical, Except:real bond returns = 1.75%real stock returns = 5.5%

24% 24% 27%

Hi t i l E tHistorical, Except:real bond returns = 0%real stock returns = 6%

47% 33% 28%

Historical, Except:l b d t 1 4% 77% 57% 46%real bond returns = -1.4%

real stock returns = 4.6%77% 57% 46%

Historical, Except:real bond returns = -1.4% for 10 Years (then 2.6% thereafter)real stock returns = 4 6% for 10 Years (then 8 6% thereafter)

43% 32% 38%real stock returns = 4.6% for 10 Years (then 8.6% thereafter)

Historical, Except:real bond returns = -1.4% for 5 Years (then 2.6% thereafter)real stock returns = 4.6% for 5 Years (then 8.6% thereafter)

22% 18% 18%

9/10/2014

16

Th 4% R lThe 4% Rule

A UniqueCause of

Sequence RiskSequence Risk

William BengenJournal of Financial Planning, October 1994

9/10/2014

17

William Bengen’s 4% RuleFigure 2.1

Maximum Sustainable Withdrawal Rates

8

9

For 50/50 Asset Allocation, 30-Year Retirement Duration, Inflation Adjustments, No FeesUsing SBBI Data, 1926-2010, S&P 500 and Intermediate-Term Government Bonds

6

7

hdra

wal

Rat

e

3

4

5

SAFEMAX

Sus

tain

able

Wit

1

2

3

Max

imum

1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 19900

Retirement Year

9/10/2014

18

Our Hapless 1966 Retiree30-year retirement period: 1966 – 199550/50 Portfolio: 4.04% was highest sustainable WR

Average 50/50 Real Return (1966-1995): 4.74%C d d 50/50 R l R (1966 1995) 4 17%Compounded 50/50 Real Return (1966-1995): 4.17%

If 1966 retiree could enjoy fixed compounded return:WR = 5 67%WR = 5.67%

Using 4.04% instead, 98% of initial real wealth remains after 30 years

9/10/2014

19


Reverse Return Sequence: 1995 back to 1966, WR 7 62%WR = 7.62%

Using 4 04% instead 160% of initial real wealth stillUsing 4.04% instead, 160% of initial real wealth still available after 30 years

9/10/2014

20

Problems with 4% Rule

•Incongruity of funding a smooth g y gspending stream from a volatile portfolio – this is a unique cause ofportfolio this is a unique cause of sequence of returns risk

•Solutions: Let spending fluctuate or lower the volatility

9/10/2014

21

Managing Sequence Risk:

Reduce Portfolio Volatility

9/10/2014

22

RelationshipBBetweenReturns,VolatilityVolatility,

&SustainableSpending

Journal of Financial Planning, January 2012 (article link)

9/10/2014

23

9/10/2014

24

9/10/2014

25

Rising Equity Glid hGlidepaths

(with Michael Kitces)(with Michael Kitces)

Journal of Financial Planning, January 2014

9/10/2014

26

Traditional Target Date Funds “T ” R tir m t“To” Retirement

100%

60%

80%

Bonds

ocat

ion

20%

40%

Stocks

Sto

ck A

llo

40+ 35 30 25 20 15 10 5 0 5 10 15 20 25 30 35 40+0%

20%

Years Before Retirement Years After Retirement

9/10/2014

27

Traditional Target Date Funds “Thr h” R tir m t“Through” Retirement

100%

60%

80%

Bonds

catio

n

20%

40%

Stocks

Sto

ck A

lloc

40+ 35 30 25 20 15 10 5 0 5 10 15 20 25 30 35 40+0%

20%


9/10/2014

28

Financial Planning: M i t i Hi h St k All tiMaintain High Stock Allocation100%

Bonds

60%

80%Bonds

loca

tion

20%

40%

Stocks

Sto

ck A

l

40+ 35 30 25 20 15 10 5 0 5 10 15 20 25 30 35 40+0%


9/10/2014

29

Target Date Funds Ri i Eq it Glid p thRising Equity Glidepath

100%

60%

80%Bonds

ocat

ion

20%

40%

Stocks

Sto

ck A

llo

40+ 35 30 25 20 15 10 5 0 5 10 15 20 25 30 35 40+0%

20%


9/10/2014

30

9/10/2014

31


Using rising glidepath, 20% stocks increasing to 50%, WR 4 06%WR = 4.06%

Less market risk still provides slightly improvedLess market risk still provides slightly improved outcome

9/10/2014

32

Managing Sequence Risk:g g q

Flexible Spending Strategies

9/10/2014

33

Spending Flexibility and the I iti l Sp di R tInitial Spending RateJonathan Guyton, Journal of Financial J y , J fPlanning, October 2004: “Is the ‘Safe’ Initial Withdrawal Rate Too Safe?”

Conclusion: Clients who are willing to make provisions for spending cuts, should the p p gneed arise, can confidently start with a higher withdrawal rate than deemed appropriate when using a constant inflation-pp p gadjusted strategy.

9/10/2014

34

Example: Adjusting Distrib tions to Market Ret rnsDistributions to Market Returns

Constant Inflation-Adjusted AmountInitially withdraw 4 3% Adjust for InflationInitially withdraw 4.3%, Adjust for Inflation

Constant PercentageA ll i hd 4 3% f R i i B lAnnually withdraw 4.3% of Remaining Balance

Guyton-Inspired Distribution Managementi i ll i hd 4 3% Adj f fl iInitially withdraw 4.3%, Adjust for Inflation

Reduce spending by 10% if WR Exceeds 5.2%Increase spending by 10% if WR Falls Below 3.4%Don’t adjust for inflation after years when Return<0%

9/10/2014

35

$10 000$11,000$12,000

FigureIndividual Simulated Paths for Spending and Wealth

For a Couple with $100,000 Wealth and a 60% Stock AllocationSimulations Based on Historical Averages

Constant AmountConstant Percent

$4,000$5,000$6,000$7,000$8,000$9,000

$10,000

With

draw

al A

mou

nt Guyton Rules

65 70 75 80 85 90 95 100 105$0

$1,000$2,000$3,000

Age

Rea

l

Spend MoreWhen ThingsGo Well

$150,000

$175,000

$200,000

ealth

Go Well...

$50,000

$75,000

$100,000

$125,000

Rea

l Rem

aini

ng W

e

65 70 75 80 85 90 95 100 105$0

$25,000

Age

9/10/2014

36

$$11,000$12,000

FigureIndividual Simulated Paths for Spending and Wealth

For a Couple with $100,000 Wealth and a 60% Stock AllocationSimulations Based on Historical Averages

Constant AmountConstant Percent

$4,000$5,000$6,000$7,000$8,000$9,000

$10,000

With

draw

al A

mou

nt

Constant PercentGuyton Rules

65 70 75 80 85 90 95 100 105$0

$1,000$2,000$3,000

,

Age

Rea

l W

But PreserveAssets By C tti S di

$150,000

$175,000

$200,000

ealth

Cutting SpendingWhen ThingsGo Poorly

$25 000

$50,000

$75,000

$100,000

$125,000

Rea

l Rem

aini

ng W

e

65 70 75 80 85 90 95 100 105$0

$25,000

Age

9/10/2014

37

Retirement Income Strategygy

Total

Process

Combine ToolsIncome Total

WealthCombine Tools

Meet Goalsfor Life

Risk Management

9/10/2014

38

Thank you! Any Questions?y y Q

Wade D. PfauWade D. PfauThe American College McLean Asset ManagementinStream Solutions

wadepfau@gmail [email protected] @WadePfau (Twitter)

wpfau.blogspot.com (Retirement Researcher Blog)

wadepfau sequence of returns risk -...

Documents