Retired Household Responses to a Persistent Low Interest ... · 5/4/2017 · May 4 2017 Retired...

Presentation to

Retired Household Responses to a Persistent Low Interest Rate Environment

Jason S. Seligman, Senior Economist, ICI

Jason J. Fichtner, Senior Research Fellow, Mercatus Center

May 4th 2017

Pension Research Council & The Boettner Center for Pensions and Retirement Research

Motivation

» interest rates have been at historic lows since 2008» some benchmark rates have in fact been negative

»we investigate how retirees are coping with these low rates» we employ HRS panel data, we target a retiree sample» We look for patterns / strategies for coping with low rates

»broadly, we find:» high wealth households have benefitted from strong equity returns» low wealth households have depleted assets including home equity

Retired Household Responses to a Persistent Low Interest Rate Environment 1May 4 2017

HRS Data

»biannual data targeting households aged 50+ from 1992-2014

»opportunity to look at self-reported wealth and allocations

»conditioning on:

»households’ placement within the 2014 wealth distribution

»controlling for:

»risk preference, home ownership & socio-economic factors.

» low interest rate period from 2008 – 2014.

Retired Household Responses to a Persistent Low Interest Rate Environment2May 4 2017

Graphical Analysis of Retirees 1992 – 2014

»we will next take six looks at retiree assets

»note the relative evolution of assets & allocations by group

»the next six slides will depict the evolution of:

» wealth

»equity & mutual fund allocations

» independent bond & safe asset allocations

» liquid assets & LTV for households’ primary residence


Wealth Evolution Among the Retired I of VI


For HRS households with a retired person; 2015 dollars ( = 2008 wave)

-100,000

0

100,000

200,000

300,000

400,000

500,000

-2 0 2 4 6 8 10 12 14 16 18 20Average Years Since Retirement

below median

bottom 10-percent

all but top 10 - percent


75 - percent

25 - percent

-100,000

900,000

1,900,000

2,900,000

3,900,000

4,900,000

5,900,000

6,900,000

7,900,000

8,900,000


top 10 - percent

top 1 - percent


For HRS households with a retired person; percentage of wealth ( = 2008 wave)

Equity & Mutual Fund Allocations II of VI

-4%

-2%

0%

2%

4%

6%

8%

10%

-2 0 2 4 6 8 10 12 14 16 18 20

Average Years Since Retirement

below median

bottom 10-percent



75 - percent

25 - percent

-5%

0%

5%

10%

15%

20%

25%

30%


top 10 - percent

top 1 - percent

Note: some portion of represented rebalancing could be automatic.



Bond Allocations III of VI

-0.1%

0.1%

0.3%

0.5%

0.7%

0.9%

1.1%


below median

bottom 10 - percent



75 - percent

25 - percent

-1%

1%

3%

5%

7%

9%


top 10 - percent

top 1 - percent




Safe Asset Allocations: CDs, Savings Bonds, T-bills IV of VI

-2%

-1%

0%

1%

2%

3%

4%

5%

6%


below median

bottom 10 - percent

25 - percent

lower 25 - percent

10 - percent

all but top:1 - percent

0%

2%

4%

6%

8%

10%

12%

14%


top 10 - percent

top 1 - percent



Proportion of Liquid to Total Wealth V of VI

For HRS households with a retired person; percentage ( = 2008 wave)

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%


below median

bottom 10 - percent

75 - percent...

25 - percent

0%

5%

10%

15%

20%

25%


top 10 - percent

top 1 - percent



For HRS households with a retired person; percentage of home’s worth ( = 2008 wave)

Loan to Value VI of VI

10%

20%

30%

40%

50%

60%

70%

80%

90%


below median

bottom 10 - percent

75 - percent

25 - percent

0%

5%

10%

15%

20%

25%

30%

-2 0 2 4 6 8 10 12 14 16 18 20


top 10 - percent

top 1 - percent

Note: it is possible to break out changes in mortgage balances and home values. home values have declined and mortgage values have increased among the bottom 10 percent of the wealth distribution

HRS Panel Data Regression Analyses

»for wealth we employ a panel spanning: 1992-2014

»for allocations we employ a panel Tobit method to handle potential censoring at zero

»the next four slides will offer a highline look at what we see


Regression – Total Wealth I of IV


In various specifications results coefficients were within a range of -0.2 to 0.5 percent of the figures above, in all cases.

Controls:

household labor force participation {0, 1}, household reports retirement {0, 1}, number of years retired, household holds a DB pension {0,1}, household holds a DC account {0,1}, household has OASI income {0,1}, household has SSI or DI income {0,1}, home ownership {0,1}, risk {1, …, 4} least to most risk averse, education {max: respondent, spouse}, education range {max - min: respondent, spouse}, whether household is white {0, 0.5, 1}, whether Hispanic {0, 0.5, 1}, respondent is female, respondent is married, married & female, age {max: respondent, spouse}, age squared, cohort controls, constant.

Panel Regression - Dependent Variable:

Total Assets ($2015) coefficient z-stat

low interest rate era {0,1} -$84,064 -10.56

2014 wealth in bottom 10 percent {0,1} -$104,999 -3.38

2014 wealth in top 10 percent {0,1} $1,233,174 60.65

general result

Equity Allocations II of IV


In various specifications results coefficients were within a range of -0.1 to 0.2 percentage point of the figures above, in all cases.

Controls:


Panel Tobit - Dependent Variable:

Equity and Mutual Fund Allocations coefficient z-stat

low interest rate era {0,1} -1.5% -5.59

2014 wealth in bottom 10 percent {0,1} -2.4% -3.73

2014 wealth in top 10 percent {0,1} 8.8% 21.89

general result

Bond Allocations III of IV


In various specifications results coefficients were within 0.1 percentage point of the figures above, in all cases.

Controls:



Bond Allocations coefficient z-stat

low interest rate era {0,1} -0.1% -4.27

2014 wealth in bottom 10 percent {0,1} -0.2% -1.92


general result

Use of Home Equity IV of IV


In various specifications results coefficients were within 1.5 percentage points of the figures above, in all cases.

Controls:



Loan to Value | Home Ownership coefficient z-stat

low interest rate era {0,1} 2.2% 7.20

2014 wealth in bottom 10 percent {0,1} 85.2% 13.40


general result

Summary of Findings

»households report declined shifts into bonds and safe assets.

»instead, we find negative wealth impacts in general

» - and that: » high wealth households have benefitted from strong equity returns

» low wealth households have depleted assets – including home equity

» on average, we observe wealth being near depletion 16 – 18 years following retirement for the bottom quarter of our sample.


Policy Implications

»given observed asset declines social insurance is acutely valuable for financing retirement

Policies should educate, encourage and facilitate claiming Social Security retirement benefits in order to maximize the utility gained from the inflation-protected annuity value offered.


Thank You

» comments welcome

Jason J. Fichtner Senior Research Fellow, Mercatus Center

[email protected]

Jason S. Seligman Senior Economist, ICI

[email protected]

Presentation DatePresentation Title Goes Here

17

Home & Mortgage Values, decoupled

May 4th 2017 Retired Household Responses to a Persistent Low Interest Rate Environment 18

Notice difference in scales

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Primary House ValueFor retired HRS Data; by 2014 wealth quantile, expressed in 2015 dollars


25

50

75

bottom 10-percent

0

10,000

20,000

30,000

40,000

50,000

60,000

-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Value of all Mortgages and Contracts (Primary Residence)For retired HRS Data; by 2014 wealth quantile, expressed in 2015 dollars


25

50

75

bottom 10-percent

Retired Household Responses to a Persistent Low Interest ... · 5/4/2017 · May 4 2017 Retired...

Documents

Transcript of Retired Household Responses to a Persistent Low Interest ... · 5/4/2017 · May 4 2017 Retired...