Copyright © 2014 by the American Academy of Actuaries

All Rights Reserved.

Social Security

Disability Insurance Trust Fund:

Behind the Numbers

Stephen C. Goss, MAAA, ASA

Chief Actuary, Social Security Administration

Moderator: Donald E. Fuerst, MAAA, FSA, FCA, EA

Senior Pension Fellow, American Academy of Actuaries

April 23, 2014

Copyright © 2014 by the American Academy of Actuaries

All Rights Reserved.

The Academy Capitol Forum: Meet the Experts

Social Security

Disability Insurance Trust Fund:

Behind the Numbers

Presentation by Stephen C. Goss, Chief Actuary,

Social Security Administration

American Academy of Actuaries

Webinar

April 23, 2014

3

Social Security Disability Insurance

155 million workers under age 66 are insured against

becoming unable to work

9 million workers now receive DI benefits

• 2 million “dependents” - mostly children

Many more protected from loss of insured status

• And from lower retirement benefits

Benefits replace 40% to 45% of career earnings on

average

• 76% for very-low earner, 27% for steady maximum earner

4

Solvency of the DI Trust Fund Reserve depletion projected for 2016 right after 1994 reallocation

Remember---the Trust Funds cannot borrow under current law

0

50

100

150

200

250

1990 1995 2000 2005 2010 2015 2020 2025

Reserv

es a

s %

of

An

nu

al

Co

st

DI Trust Fund Ratio in 1995 Trustees Report

Tax-Rate Reallocation

in 1994

5

Solvency of the DI Trust Fund looked MUCH better in 2007 Boost from the “new economy” anticipating NO recession

DI Trust Fund Ratio in 1995 and 2008 Trustees Reports

0

50

100

150

200

250

1990 1995 2000 2005 2010 2015 2020 2025

Re

se

rve

s a

s %

of

An

nu

al

Co

st

1995TR

2008TR

"New Economy"

6

Solvency of the DI Trust Fund; reserve depletion in 2016 2008 recession offset “new economy”; cycles still happen

DI Trust Fund Ratio in 1995, 2008, 2013 Trustees Reports

0

50

100

150

200

250

1990 1995 2000 2005 2010 2015 2020 2025

Reserv

es a

s %

of

An

nu

al

Co

st

1995TR

2008TR

2013TR"New Economy"

irrational

exuberance

2008

Recession

back to

reality

7

Economic cycles and policy changes fluctuate,

and DI incidence rates also vary

0

1

2

3

4

5

6

7

8

9

10

11

12

1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020

Un

em

plo

yme

nt

rate

an

d D

isa

ble

d w

ork

er

incid

en

ce

per

tho

usa

nd

exp

os

ed

Calendar year

Age-sex-adjusted disabled worker incidence rate

Civilian unemployment rate

Historical Estimated

Recession and SSI

1974

Recession

Recession

Recession

1980 Amendments: PER, CDRs, EPE,

Lowered Family Max

1984 Amendments: Multiple Impairments Medical Improvement

Mental Listings

1996 Amendments: Drug Addiction & Alcohol

CDR Plan 1996-2002

SSI Outreach

1970-74 Large Benefit

Increases

Recession

8

Most of the recession effect is from

less GDP, not more DI cost

Change in DI Benefit Cost and in GDP

Between 2008 TR and 2013 TR

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

2010 2011 2012 2013

Increase in DIBenefit Cost

Reduction in GDP

Increase in DIBenefit Cost/GDP

9

Additional disabled worker beneficiaries are a

small fraction of reduced employment

Changes in Disabled Worker Beneficiaries and in

Covered Workers from 2008 TR to 2013 TR

0

2,000

4,000

6,000

8,000

10,000

12,000

2007 2008 2009 2010 2011 2012 2013

Ch

an

ge

in

Th

ou

sa

nd

s

Increase inDisabledWorkers

Reduction inCoveredWorkers

10

Is DI out of control, taking over OASDI? (Note 5% increase in DI cost for 2010 due to recession)

0

2

4

6

8

10

12

14

16

18

20

1980 2010 2040

DI Cost as a Percent of Total OASDI Cost

1995 TR

2013TR

12.8

16.917.9

12.6 12.312.8

11

DI cost as percent of GDP has peaked,

but scheduled income is too low

DI Cost and Income as Percent of GDP 1975-2090 2013 Trustees Report Intermediate Assumptions

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

2055

2060

2065

2070

2075

2080

2085

2090

Baby Boomers

reach ages 45-64

in 2010

Baby Boomers

reach ages 25-44

in 1990

DI Cost

DI Income

Note:

Recession raised

DI Cost/GDP

by 15% for 2010

Disabled workers increased 187% from 1980 to 2010;

let’s work backwards and explain

12

Population age 20-64 increased 41% from 1980 to 2010;

let’s adjust that out

13

Population age 20-64 is much older in 2010 Boomers have aged with lower-birth-rate generations following

14

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000DI Disabled Worker Beneficiaries: from 2010 to 1980, in thousands

187 percent above 1980

Age 20-64 popincreases 41%

increases 38%

15

Remarkable changes in age distribution Progression of the boomers and drop in birth rates dominate

Figure 2: Age Distribution of the Population Age 25+, 1940 to 2100

(2012TR)

0

10

20

30

40

50

60

70

80

90

100

1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Perc

en

t o

f P

op

ula

tio

n a

t A

ges

25

+

25-44

25-44

45-64

65-84

85+

Boomers

become

25-44

Boomers

become

45-64

Boomers

become

65-84

The Normal Retirement Age increased from 65 to 66,

adding 4% more disabled worker beneficiaries

16

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000DI Disabled Worker Beneficiaries: from 2010 to 1980, in thousands

187 percent above 1980

Age 20-64 popincreases 41%

increases 38%

increases 4%

17

Increased work by women raised insured;

men a little lower at younger ages

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%Figure 5: Percent of Population that is Insured for Disability

Male

Female

Disability insured rates in the population increased

substantially for women, mainly at higher ages;

increased beneficiaries by 21%

18

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000DI Disabled Worker Beneficiaries: from 2010 to 1980, in thousands

187 percent above 1980

Age 20-64 popincreases 41%

increases 38% increases

4% increases 21%

Recession of 2008-10 increased disabled workers 5%

compared to full-employment economy,

as had been experienced prior to 1980

19

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000DI Disabled Worker Beneficiaries: from 2010 to 1980, in thousands

187 percent above 1980

Age 20-64 popincreases 41%

increases 38% increases

4% increases 21%

increases 5%

This leaves 12% increase for all other causes; the

increase in disability incidence rates for women easily

explains this

20

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000DI Disabled Worker Beneficiaries: from 2010 to 1980, in thousands

187 percent above 1980

Age 20-64 popincreases 41%

increases 38%

increases 4%

increases 21%

increases 5%

increases 12%

IncidenceRates, etc

21

Incidence rates for women have risen

to male level Figure 8: New Disabled Workers per 1,000 Exposed (Incidence)

Age-Adjusted (2000) - 2012 Trustees Report

2

3

4

5

6

7

8

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

2055

2060

2065

2070

2075

2080

2085

2090

New

Aw

ard

s p

er 1

,000

Exp

ose

d

Male

Female

But NOT because of increasing mental impairment

for young females: steady distribution by impairment

Figure 12: Female Age 30-39 disabled worker new entitlement distribution by primary diagnosis (awarded through June 2012)

22

Nor for young males: note steady but for HIV

bulge in 1986-2000 Figure 13: Male Age 30-39 disabled worker new entitlement distribution by primary diagnosis (awarded through June 2012)

23

For older females: increased musculoskeletal

impairment; diminished circulatory Figure 14: Female Age 50-59 disabled worker new entitlement distribution by primary diagnosis (awarded through June 2012)

24

Same for older males: increased musculoskeletal

impairment; less circulatory Figure 15: Male Age 50-59 disabled worker new entitlement distribution by primary diagnosis (awarded through June 2012)

25

26

So where are we on DI? • Is the sky falling, cost out of control? No.

• Or are we following a path foreseen? Yes.

• Trust Fund reserves projected to deplete 2016 Need change soon to avoid inability to pay in full & on time

Default: Revenue enough to pay 80% of benefits, so: 1. Cut all DI benefits by 20%?

2. Increase DI tax revenue by 25%?

3. Or, reallocate tax rate between OASI and DI?

• Need further changes for long-range solvency

27

Potential tax rate reallocation between OASI and DI:

Like in 1994—NO change in total taxes

28

Some changes specific to DI • Actuarial deficit for DI is 0.32 percent of payroll

– Changes considered by Senator Coburn in 2011 http://www.ssa.gov/OACT/solvency/TCoburn_20110718.pdf

• Raise ages for vocational factors by up to 8 years

– Lowers actuarial deficit by 0.04 percent of payroll

• Eliminate “reconsideration” level of disability appeal

– Increases actuarial deficit by 0.02 percent of payroll

• Close record without exception after first ALJ decision

– Must reapply with new evidence

– Lowers actuarial deficit by 0.01 percent of payroll

• Time limit benefits: MIE 2 years, MIP 3 years, MINE 5 years

– Reapply; may deny without medical improvement

– Lowers actuarial deficit by 0.10 percent of payroll

29

Withhold DI when receiving

Unemployment Insurance payments

• Currently no DI offset for receiving UI

• Change considered by Representative Johnson in 2013 http://www.ssa.gov/OACT/solvency/SJohnson_20140107.pdf

– Treat any month with UI payment as SGA

– Lowers actuarial deficit by 0.01 percent of payroll

• Change considered by Senator Coburn in 2013 http://www.ssa.gov/OACT/solvency/TCoburn_20140107.pdf

– Suspend DI benefit for any month with UI payment

– Lowers actuarial deficit by 0.01 percent of payroll

• Another possibility–offset DI benefit dollar for dollar for UI

30

Changes for long-range DI solvency

• Actuarial deficit for DI is 0.32 percent of payroll – Need to lower DI cost 20% or increase DI revenue 25%

– Or, some combination of these

• Will likely be addressed in overall OASDI changes – Note that increasing NRA shifts cost to DI

– May need further tax rate reallocation to DI in final amendments

• For overall OASDI solvency: http://www.ssa.gov/oact/solvency/provisions/index.html

• Increase tax rate or raise/eliminate the taxable maximum

• Lower the benefit (PIA level)

• Increase the NRA

• Expand the tax base – Cover all state and local government employees

– Tax employer-sponsored group health insurance premiums

31

Finally A Little Extra

Credit

32

Remember DI is Just Part of Social Security

• Any fix for the long-term will have to be

comprehensive

– Address the “Aging” of the population

• “Macro Aging” Shift toward more elders, because

Slowed growth for younger ages

Faster growth for older ages

• “Micro Aging” People are living longer

Lower death rates

Higher life expectancy dd

33

Changing age distribution over next 20 years

mainly due to Macro Aging – a permanent level shift

Age Distribution of the Population Age 25+, 1940 to 2100 (2012TR)

0

10

20

30

40

50

60

70

80

90

100

1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Perc

en

t o

f P

op

ula

tio

n a

t A

ges 2

5+

25-44

25-44

45-64

65-84

85+

Boomers

become

25-44

Boomers

become

45-64

Boomers

become

65-84

34

The level shift in age distribution is NOT

due to a sudden shift in life expectancy

35

Why so much “Macro Aging”? Birth rates. If birth rates had stayed at 3.0 per woman after the “boom”?

Age Distribution of the Population Age 25+, 1940 to 2100 (2012TR):

What If Birth Rate (TFR) Had Stayed at 3.0?

0

10

20

30

40

50

60

70

80

90

100

1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Perc

en

t o

f P

op

ula

tio

n a

t A

ges 2

5+

25-44

45-64

65-84

85+

Boomers

become

25-44

Boomers

become

45-64

Boomers

become 65-84

TFR=Actual &

TR intermediate

TFR remains at 3.0

after 1964

36

If birth rates had stayed at 3.0 or 3.3 per woman after

1964, our Aged Dependency ratio would not SHIFT

Aged Dependency Ratio (Population 65+/20-64) 2012 TR

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Actual and TR Intermediate

TFR remain at 3.0 after 1964

TFR remain at 3.3 after 1964

37

Even if birth rates returned to 3.0 or 3.3 per woman after 2014,

our Aged Dependency ratio would come back down

Aged Dependency Ratio (Population 65+/20-64) 2012 TR

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Actual and TR Intermediate

TFR return to 3.0 after 2014

TFR return to 3.3 after 2014

38

BUT birth rates are not going back up in the U.S. They are staying around 2.0 TFR, high among developed nations

U.S. Total Fertility Rate: With and Without Adjustment for Survival to Age 10

0

1

2

3

4

5

1875 1885 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005

Ave TFR Ave AdjTFR

1875-1925 3.67 2.85

1926-1965 2.84 2.69

1966-1990 1.99 1.95

1991-2003 2.01 1.99

TFR

AdjTFR

39

So we need to address a level shift in cost that is mainly

due to lower birth rates and not due to greater longevity

U.S. Social Security Cost and Income as percent of GDP

40

Implications for Social Security

–The older age distribution requires:

• Beneficiaries receive less--- 25%

less,

• Workers pay more--- 33%

more,

• Increase “Normal Retirement Age”---7+

yrs,

• Or some combination

Questions?

41