CBO’s Social Security Projections (presentation)

cbo 666 views 21 slides May 15, 2025
Slide 1
Slide 1 of 21
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21

About This Presentation

Presentation by Julie Topoleski, CBO’s Director of Labor, Income Security, and Long-Term Analysis, at the 17th Annual Meeting of the OECD Working Party of Parliamentary Budget Officials and Independent Fiscal Institutions.


Slide Content

Presentation at the 17th Annual Meeting of the OECD Working Party of
Parliamentary Budget Officials and Independent Fiscal Institutions
May 15, 2025
Julie Topoleski
Labor, Income Security, and Long-Term Analysis Division
CBO’s Social Security Projections
For more information about the Organisation for Economic Co-operation and Development (OECD) and this meeting, see https://tinyurl.com/2c4xskkf.

1
Social Security—the largest single program in the federal budget—has two
components:
§Old-Age and Survivors Insurance (OASI) provides benefits to retired workers,
their eligible dependents, and some survivors of deceased workers.
§Disability Insurance (DI) provides benefits to disabled workers and their
dependents.
The program is financed by payroll taxes and income taxes on benefits.
Revenues from those taxes are credited to the OASI and DI trust funds.
Benefits are based on a progressive benefit formula and average earnings.
The U.S. Social Security System

2
CBO’s 10-Year Social Security
Baseline and Cost Estimates

3
Total benefit outlays depend on the number of beneficiaries and their average
benefits.
The number of beneficiaries grows with the relevant population, adjusted for
changes in the share of a specific age–sex group receiving benefits.
Average benefits grow primarily with inflation, and to a lesser extent, with
economywide wage growth.
Cost estimates are often for small subgroups that are not modeled separately in
the baseline. For those estimates, CBO develops more detailed projections to
supplement the baseline.
CBO’s baseline projections reflect the assumption that benefits will be paid as
scheduled under the provisions of the Social Security Act, regardless of the
balances of the trust funds.
Ten-Year Baseline and Cost Estimates

4
For more information and for notes about this table, see the online version at https://tinyurl.com/3xx9v2hw. For the equivalent table for Disability Insurance,
see https://tinyurl.com/bdzn9kyn.
Information About OASI in CBO’s 10-Year Baseline
202420252026202720282029203020312032203320342035
BUDGET INFORMATION
Billions of dollars, by fiscal year
Estimated outlays for benefits1,293.81,405.01,486.71,573.01,669.01,764.71,861.91,962.02,063.52,165.82,267.62,370.6
Components of the Social Security
Old-Age and Survivors Insurance Program
Benefit outlays
Retired workers and families
Retired workers1,090.71,182.91,259.81,340.21,429.11,518.01,608.61,702.11,796.81,891.61,985.62,080.1
Wives and husbands35.640.841.742.944.546.047.548.950.351.653.054.4
Children 7.88.18.48.89.29.610.010.310.711.111.411.7
Survivors
Children 28.329.229.930.731.532.233.033.834.635.336.136.9
Mothers and fathers1.71.71.71.61.61.61.61.61.61.51.51.5
Aged widows and widowers127.2139.8142.6146.2150.7154.8158.8162.8167.1172.0177.5183.4
Disabled widows and widowers2.32.32.32.32.22.22.22.22.32.32.32.3
Parents ************
Lump-sum death benefit0.20.20.20.20.20.20.20.20.20.20.20.2

5
For more information and for notes about this table, see the online version at https://tinyurl.com/3xx9v2hw. For the equivalent table for Disability Insurance,
see https://tinyurl.com/bdzn9kyn.
Information About OASI in CBO’s 10-Year Baseline (Continued)
202420252026202720282029203020312032203320342035
PARTICIPATION AND BENEFIT INFORMATION
Beneficiaries on December 31
(thousands of people)
Retired workers and families
Retired workers51,83353,37954,88356,45757,92559,30860,60061,78662,85763,72964,46165,098
Wives and husbands1,8231,7851,7621,7421,7141,6871,6631,6401,6161,5901,5651,538
Children 695703710719726733737740742741737736
Survivors
Children 2,0322,0242,0162,0102,0041,9971,9891,9801,9691,9591,9491,941
Mothers and fathers1049994908683797672696664
Aged widows and widowers3,4833,4723,4353,4073,3543,2973,2313,1633,1133,0833,0693,062
Disabled widows and widowers195187178170163158154151148144141138
Parents 11111111********
Total beneficiaries60,16661,65063,07964,59665,97367,26468,45469,53770,51771,31571,98872,577

6
For more information and for notes about this table, see the online version at https://tinyurl.com/3xx9v2hw. For the equivalent table for Disability Insurance,
see https://tinyurl.com/bdzn9kyn.
Information About OASI in CBO’s 10-Year Baseline (Continued)
202420252026202720282029203020312032203320342035
Average monthly benefit on December 31 (dollars)
Retired workers and families
Retired workers1,9772,0632,1402,2142,2902,3682,4492,5362,6252,7182,8142,917
Wives and husbands9261,0121,0381,0631,0951,1261,1591,1921,2281,2641,3011,341
Children 9179459761,0041,0331,0631,0951,1291,1641,2001,2381,278
Survivors
Children 1,1331,1651,1991,2301,2651,2921,3251,3601,3981,4291,4651,503
Mothers and fathers1,3151,3551,3981,4361,4761,5161,5571,6011,6451,6901,7351,784
Aged widows and widowers1,8331,9642,0192,0722,1342,1982,2652,3352,4062,4802,5592,640
Disabled widows and widowers9529791,0081,0331,0601,0871,1151,1451,1741,2041,2351,268
Parents 1,6771,7401,8071,8711,9382,0052,0762,1502,2262,3042,3852,471

7
For more information and for notes about this table, see the online version at https://tinyurl.com/3xx9v2hw. For the equivalent table for Disability Insurance,
see https://tinyurl.com/bdzn9kyn.
Information About OASI in CBO’s 10-Year Baseline (Continued)
202420252026202720282029203020312032203320342035
KEY ASSUMPTIONS
By calendar year
Average wage for indexing (dollars)70,36272,94575,17978,08881,11184,09987,06390,06593,13896,32799,633103,055
Taxable maximum (dollars)168,600176,100186,000192,600198,600206,400214,200222,300230,100237,900246,000254,400
PIA for hypothetical retired worker with
average earnings (equal to AWI) (dollars)a2,4442,5502,6952,7882,8742,9823,0983,2123,3253,4393,5573,682
PIA for hypothetical retired worker with
maximum earnings (dollars)a 3,9454,1184,3534,5054,6464,8235,0115,1975,3805,5675,7575,957
COLA (percent)b 2.52.42.52.32.32.22.22.22.22.22.22.3
First month for which the COLA is effective12/202412/202512/202612/202712/202812/202912/203012/203112/203212/203312/203412/2035

8For more information and for notes about this table, see the online version at https://tinyurl.com/435cm8tf.
Information About the Social Security Trust Funds in
CBO’s 10-Year Baseline
202420252026202720282029203020312032203320342035
BUDGET INFORMATION
Billions of dollars, by fiscal year
Income to trust funds
Payroll tax revenue1,2601,3041,3641,4181,4721,5301,5911,6541,7181,7821,8481,917
Other income
Income taxes on benefits5467828994100106112118125133140
Federal payroll tax, employer sharea232324252627272829303132
Interesta 676862615854484130202225
Subtotal, other income144158168176179181181181177175186196
Total income1,4041,4621,5321,5941,6501,7111,7721,8351,8951,9572,0342,113
Outgo from trust funds
Benefits1,4481,5661,6581,7541,8591,9612,0652,1732,2822,3932,5042,617
Discretionary administration667777788888
Treasury administration111111111111
Railroad Retirement transfer666666666666
Total outgo1,4611,5781,6711,7681,8721,9752,0792,1872,2972,4082,5192,632
Net cash flow (includes interest income)
Old-Age and Survivors Insurance-91-154-177-211-261-306-352-401-454-506-543-580
Disability Insurance353838373942464953565861
Primary cash flow (excludes interest income)
Old-Age and Survivors Insuranceb-154-215-231-262-309-347-386-426-466-506-543-580
Disability Insurance303130282829313334353636
End-of-year balance
Old-Age and Survivors Insuranceb2,5822,4282,2512,0401,7791,4731,121719265ccc
Disability Insurance178216254291330372418467520575634694

9
CBO’s 75-Year Social Security
Projections

10Congressional Budget Office, An Overview of CBOLT: The Congressional Budget Office Long-Term Model (April 2018), www.cbo.gov/publication/53667.
The model was created in the early 2000s.
§First, CBO secured a statutory right to administrative microdata.
§The initial model was a cell-based model similar to the model used by the
Social Security trustees.
§Once CBO obtained microdata, a team of analysts developed a
microsimulation model.
The microsimulation model starts with information drawn from the Social Security
Administration’s continuous work history sample (CWHS)—a 1percent sample of
people with Social Security numbers.
§The CWHS includes data on people’s sex, birth year, earnings, and type and
amount of Social Security benefits.
§It does not provide all of the information about people that the microsimulation
model needs.
CBO’s Long-Term Social Security Model

11Congressional Budget Office, An Overview of CBOLT: The Congressional Budget Office Long-Term Model (April 2018), www.cbo.gov/publication/53667.
The microsimulation model projects annual demographic changes and economic
outcomes for each person in the representative sample. Those changes and
outcomes include:
§Marriage
§Death
§Emigration
§Labor force participation
§Unemployment
§Earnings
§Claiming of Social Security benefits
The model is longitudinal, which allows for consistency between multiple
outcomes for each person in the model.
CBO’s Long-Term Social Security Model (Continued)

12
Outcomes Projected by CBO’s Model

13Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Social Security’s Outlays and Revenues, With Scheduled and
Payable Benefits
In a scheduled-benefits
scenario, the Social Security
Administration continues to
pay benefits as scheduled
under current law, regardless
of the status of the program’s
trust funds.
In a payable-benefits
scenario, after the trust funds’
balances were exhausted,
revenues would be
insufficient to pay the full
benefits specified in law. In
that scenario, CBO assumes
that annual outlays for Social
Security would be limited to
the annual revenues credited
to the program.

14
Ratio of trust fund balance to scheduled payments
Old-Age and
Survivors Insurance
Disability
Insurance
Combined
Old-Age,
Survivors, and
Disability
Insurance
Projected
0
1
2
3
4
5
206520602055205020452040203520302025202020152010200520001995
Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Social Security Trust Fund Ratios
A trust fund’s balance is
exhausted when it reaches
zero. In CBO’s projections,
the balance of the OASI
trust fund is exhausted in
fiscal year 2033, and the
balance of the DI trust fund
is exhausted in 2064. If the
balances of the OASDI trust
funds were combined, they
would be exhausted in fiscal
year 2034.

15Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Summarized 75-Year Financial Measures for Social Security,
With Scheduled Benefits
The income rate for Social
Security measures the
program’s income from
taxes as a percentage of
gross domestic product or
taxable payroll over a given
period. The cost rate
measures the program’s
cost as a percentage of
GDP or taxable payroll.
As a percentage of GDPAs a percentage of taxable payroll
Trust fund
Income
rateCost rate
Actuarial
balance
(difference)
Income
rateCost rate
Actuarial balance
(difference)
Old-Age and Survivors Insurance4.15.5-1.411.916.1-4.2
Disability Insurance0.60.7*1.82.0-0.1
Old-Age, Survivors, and Disability
Insurance4.76.2-1.513.818.1-4.3
* = between zero and 0.05 percent.

16
2024 dollars
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
Lowest quintile Middle quintile Highest quintile
1950s
1970s
1990s
0
10
20
30
40
50
60
70
Lowest quintile Middle quintile Highest quintile
1950s 1970s
1990s
Average Initial Annual Benefits for Retired Workers, by Birth Cohort and Earnings Quintile
Percent
Average Initial Replacement Rates for Retired Workers, by Birth Cohort and Earnings Quintile
Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Initial Distributional Measures
CBO projects the benefits
people receive in their first
year of claiming them as
well as how those benefits
compare to people's
preretirement earnings that
those benefits replace.
The agency's published
report highlights the lowest,
middle, and highest
quintiles, but CBO also
publishes supplemental data
for all quintiles, for the
intervening cohorts, and for
men and women.

17
Percent
Average Lifetime Social Security Benefits as a Percentage of Lifetime Earnings, by Birth Cohort and Earnings Quintile
Percent
Average Lifetime Social Security Taxes as a Percentage of Lifetime Earnings, by Birth Cohort and Earnings Quintile
0
5
10
15
20
25
30
35
Lowest quintile Middle quintile Highest quintile
1950s
1970s
1990s
0
2
4
6
8
10
12
14
Lowest quintile Middle quintile Highest quintile
1950s
1970s 1990s
Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Lifetime Distributional Measures
The agency also projects
beneficiaries' lifetime
benefits and the Social
Security taxes that they and
their employers pay as
percentages of beneficiaries'
earnings over their working
lives.

18
Ratio
0
1
2
3
Lowest quintile Middle quintile Highest quintile
1950s
1970s
1990s
Congressional Budget Office, CBO’s 2024 Long-Term Projections for Social Security (August 2024), www.cbo.gov/publication/60392.
Lifetime Distributional Measures (Continued)
CBO then compares those
percentages to determine
the ratio of benefits people
receive to the taxes they pay.
Ratio of Average Lifetime Social Security Benefits to Average Lifetime Taxes Paid, by Birth Cohort and Earnings Quintile

19Congressional Budget Office, An Overview of CBOLT: The Congressional Budget Office Long-Term Model (April 2018), www.cbo.gov/publication/53667.
Some analyses require CBO to use other models in addition to the
microsimulation model. For example, if Social Security benefits were limited to
the amounts payable from the OASDI trust funds, benefits would have to be
reduced by about one quarter; that reduction would significantly lower federal
deficits.
§For this analysis, CBO generally assumes that benefits would be cut by an
equal percentage for all beneficiaries.
§The microsimulation model projects baseline outcomes and is used to
determine the reduction in benefits needed to keep outlays equal to revenues.
§That information is passed to CBO’s overlapping generations model. The
model incorporates the assumption that people are surprised by the benefit
reduction and thus do not alter their behavior in anticipation of the change.
§CBO also uses a multiplier model to capture aggregate demand effects in the
short term and a Solow-style growth model to capture the effects of crowding
out in the long term.
Models for Additional Analysis

20
Byoung Hark Yoo, Conditional Forecasting With a Bayesian Vector Autroregression, Working Paper 2023-08 (Congressional Budget Office, November 2023),
www.cbo.gov/publication/59629; U. Devrim Demirel and James Otterson, Quantifying the Uncertainty of Long-Term Economic Projections, Working Paper 2022-07 (Congressional
Budget Office, April 2022), www.cbo.gov/publication/57711; and Congressional Budget Office, Quantifying Uncertainty in the Analysis of Long-Term Social Security Projections
(November 2005), www.cbo.gov/publication/17472.
The results that CBO typically presents in its long-term Social Security analysis
rely on a single set of demographic and economic projections that reflect central
estimates for a number of variables.
In the past, the agency presented results based on a Monte Carlo analysis for a
set of demographic and economic variables. That analysis relied on annual
stochastic shocks applied to fixed central tendencies.
CBO has since developed new methods for assessing the uncertainty of
long-term economic projections.
§One method quantifies the uncertainty of economic variables by using
simulations from a multivariate statistical model in which variables are
formulated as sums of unobserved stationary and nonstationary components.
§CBO has also developed a Bayesian vector autoregression model to generate
sets of alternative economic projections.
Methods for Assessing Uncertainty