Adequacy of Social Security in Meeting Consensus Goals

By Paul Donahue

Retirement Section News, October 2023

Social Security is the program of taxes and benefits created by the Social Security Act of 1935, signed into law by President Franklin D. Roosevelt on Aug. 14, 1935. It was one of the signature and defining achievements of the New Deal. The monthly payments to qualifying older Americans that define the term Social Security for most Americans, and the taxes that are dedicated to their support, are the only parts of the OASDI’s (Old Age, Survivors and Disability Insurance) array of benefits and funding that we will discuss in this essay.

Although as enacted, Social Security contemplated current funding of future benefits,[1] Social Security is almost entirely a program where current benefits are funded by current taxes. In 2022, Social Security taxes[2] amounted to US$993 billion, interest provided US$64 billion, and general tax transfers provided US$183 billion. Benefit payments totaled US$1,088 billion.[3] Considered on its own, the Old Age and Survivors Insurance Trust Fund can pay full benefits only until 2033.[4]

Individual Taxpayer Return on Investment Perspective

We shall have more to say about this later, but the political mantra about Social Security pension benefit recipients: “They earned their benefits!” is largely true. The average historical return on 401(k) participant balances exceeds the rate of return for all current retirees of whatever income level.[5] If participant FICA and SECA contributions had been invested in the average 401(k) portfolio, participants would have had benefits exceeding those provided by Social Security.

Social Security as Social Insurance

The definition of social insurance, included in a presentation by Oliver Wyman for the California Department of Insurance, is a useful point of departure for this article.

Social Insurance: 1. Protection of the individual against economic hazards (such as unemployment, old age, or disability) in which the government participates or enforces the participation of employers and affected individuals; 2. Social insurance encompasses broad-based systems that help workers and their families pool risks to avoid loss due to retirement, death, disability, or unemployment; and to ensure access to health care.[6]

The following description of the characteristics of social insurance found in the Actuarial Standards Board’s Actuarial Standard of Practice No. 32 “Social Insurance” is also useful:

Characteristics of Social Insurance—Three characteristics of social insurance programs are of fundamental importance to the analysis of their actuarial status. First, because participation is essentially mandatory, social insurance programs can be assured of new entrants. Second, because such programs are operated by governments, program termination is usually not an important consideration when determining the program’s actuarial status. Third, social insurance is based on laws and regulations that can be changed (e.g., taxes or premiums may be increased or benefits decreased) without the consent of the participants.[7]

Social insurance programs are an economically efficient means of providing the designed level of protection because:

(1) Standardized, large scale administration provides scale efficiencies, and

(2) the costs of education and promotion are smaller than for private insurance programs.

Appropriate Criterion for Evaluating the Success of Social Security

A person’s evaluation of Social Security’s performance is grounded in what that person thinks Social Security ought to be doing. At one extreme, some believe that Social Security should be doing nothing at all and their proposal for reform is phasing it out.[8]

In this essay, the standard by which I am going to evaluate Social Security is this: Does Social Security provide income sufficient for a person who has worked for many years to have a decent standard of living? I do not make income redistribution a goal in itself,[9] but I do assume that some level of income redistribution is necessary to achieve the “decent standard of living” goal.[10] Also, I do not attempt any economic quantification of “decent standard of living,” but rely on older Americans’ “life satisfaction” to approximate it. My assumption is that those who are satisfied with their quality of life would say they enjoy a decent standard of living.

I realize that my statement of the goal is not particularly well-defined, but in view of the available empirical evidence, I believe that it will do for the purposes of this article.

Social Security and Retiree Income

Social Security provides more than half of income for more than half of the population aged 65 or older. It provides more than 90% of income for one quarter of that population.[11] Current estimates are that 9% of older Americans live in poverty; without Social Security an estimated 38% would live in poverty.[12] Social Security benefits are much lower in the United States than in many other countries in the developed world, at about the 40th percentile (e.g., below Turkey, Estonia and Latvia).[13]

Older Americans’ Life Satisfaction

Péter Hudomiet, Michael D. Hurd and Susann Rohwedder’s important paper “The Age Profile of Life-Satisfaction After Age 65 in the U.S.”[14] provides a wealth of valuable data for those who wish to make policy recommendations about Social Security. The principal point of the study is to explain the counterintuitive result that life satisfaction increases with age beyond retirement into advanced old age, the “paradox of well-being.”[15]

The result I found most relevant for the purpose of this essay is the distribution of average life satisfaction by income quartile. The possible survey responses ranged from 1 (not at all satisfied), 2 (not very satisfied), 3 (somewhat satisfied), 4 (very satisfied) and 5 (completely satisfied). Here are the results:

Income quartile Distribution Average Life Satisfaction
Lowest 25.0 3.76
2 25.0 3.85
3 25.0 3.94
Highest 25.0 4.03
All 100.0 3.91

The average life satisfaction for the lowest quartile was closer to “very satisfied” than to “somewhat satisfied,” and only 7% lower than that for the highest income quartile. This is consistent with the fact that the lowest income quartile from the preceding section consisted of individuals for whom Social Security provided 90% or more of their total income.

While these data sets equivalences I have implied are by no means conclusive, and while “very satisfied” does not correspond to my “decent standard of living” statement of the goal for Social Security, nevertheless we have come far enough that I am prepared to conclude that the existing Social Security system achieves its social insurance goal.

Funding Challenges

Like a new defined benefit pension plan that provides benefits for prior service, Social Security began with a significant unfunded past service cost, and benefit increases have only made it larger. The Actuarial Note “Unfunded Obligation and Transition Cost for the OASDI Program”[16] points out that in the context of the actual program’s intent and practice, it does not make sense to speak of an “unfunded liability.” However, they do describe a theoretical “transition cost.” In 2004, the “transition cost” was 13 trillion dollars, and is about 50% higher now.[17]

Rate of Return Evaluation of Social Security Taxes and Benefits

The chart below, taken from “Cohort-Specific Measures of Lifetime Social Security Taxes and Benefits,”[18] explains why making the case for cutting benefits is very difficult politically.

OASI Chart
OASI Real Internal Rate of Return, Expressed as a Percent, by Policy and Cohort

OASI Real Internal Rate of Return, Expressed as a Percent, by Policy and Cohort

Under any plausible scenario for adjustment of taxes and/or benefits needed to sustain Social Security, the rate of return on contributors’ taxes is relatively low, far below the long-term rate of return on equity, 5.86%.[19] Yet, Social Security has reduced the poverty level among the elderly by 27%.

The political obstacles to cutting benefits posed by the mantra “they earned their benefits” reinforces my belief that cutting benefits is not the right solution to the Social Security funding challenge.

Progressivity of the Current System

The internal rate of return analysis can be extended to analyze return by discrete levels of taxes paid and benefits received. The purely illustrative Chart 1 below tracks cumulative benefits paid (y-axis) and cumulative taxes paid (x-axis).[20] As a specific example, those who have paid the first 40% of Social Security taxes, who receive under a progressive tax regime (red line) 60% of the benefits paid, under a neutral regime (dotted line) 40% of the benefits paid, and under a regressive regime (blue line) 15% of the benefits paid.

Chart 1
Progressivity of Social Security: Illustrative Examples Using Stylized Lorenz Curves

Progressivity of Social Security: Illustrative Examples Using Stylized Lorenz Curves

The authors conclude: “Results show that the Social Security program is modestly progressive on a lifetime basis, about halfway between a pure DC program and a program paying a flat-dollar benefit. In this way, the current program can be described as balancing its twin goals of income adequacy and individual equity.”[21]

Conclusion

In concluding, I endorse the conclusion of Biggs et al., quoted immediately above, that the current Social Security program does a reasonable job of balancing adequacy and individual equity, but repeat my own view that it barely achieves adequacy: Reforming Social Security should mean increasing funding, not cutting benefits.

Looking Ahead

In forthcoming articles, I shall look to European efforts at reform (and some of the pitfalls along the way) to attempt to gain insight into a program of US Social Security reform that is both sustainable and achievable.

Statements of fact and opinions expressed herein are those of the individual authors and are not necessarily those of the Society of Actuaries, the newsletter editors, or the respective authors’ employers.


Paul Donahue, FSA, is a lawyer and actuary who works in New York City. Paul can be contacted at pauljdonahuefsa@aol.com.


Endnotes

[1] Social Security Act (1935), Section 201.

[2] In “taxes” I include both the OASI component of payroll taxes, 5.3% for both employers and employees (10.6% for the self-employed), and income taxes on Social Security benefits, and excluded the disability insurance portion of the taxes (0.9% for both employers and employees, and 1.8% for the self-employed).

[3] Old Age and Survivors Insurance Trust Fund, https://www.ssa.gov/oact/STATS/table4a1.html.

[4] Research, Statistics and Policy Analysis, Summary: Actuarial Status of the Social Security Trust Funds, (March, 2023).

[5]According to Vanguard’s invaluable How America Saves (2022), the average 401(k) participant had 77% invested in equity. The average participant through age 64 had at least 60% in equity. If we combine data about historical rates of return with data on historical inflation, the real rate of return on an asset allocation of 60% equity, 40% fixed from 1926 to 2019 was 5.86% (8.77 – 2.91). Cf. Vanguard, Historical Index risk/return (1926-2019). https://advisors.vanguard.com/VGApp/iip/advisor/csa/analysisTools/portfolioAnalytics/historicalRiskReturn , and CPI Inflation Calculator, https://www.in2013dollars.com/1926-dollars-in-2019?amount=410. Compare that to the rates of return for accumulated payroll taxes in the OASI chart.

[6] Social Insurance vs. Public Assistance, Oliver Wyman, July 16,2021, chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/http://www.insurance.ca.gov/0500-about-us/03-appointments/upload/SocialInsuranceVsPublicAssistancePresentation20210716.pdf

[7] American Academy of Actuaries, Adopted January 1998, Updated Effective May 31, 2011.

[8] The Libertarian Party platform states: “Retirement planning is the responsibility of the individual, not the government. Libertarians would phase out the current government-sponsored Social Security system and transition to a private voluntary system.” https://www.lp.org/platform/ The extent to which Republicans have in the recent past favored extreme cutbacks is a matter of dispute. Cf. “What Republicans have actually said about cuts to Social Security and Medicare.” Julia Shapiro, The Hill, Feb. 8, 2023. https://thehill.com/business/3850036-republicans-social-security-medicare-debt-ceiling/

[9] Nor do I make avoiding income redistribution to the maximum extent possible, which is my understanding of what some commentators mean by “individual equity” a goal in itself.

[10] Andrew Bigss, “A New Vision for Social Security,” summer 2013, https://www.nationalaffairs.com/publications/detail/a-new-vision-for-social-security, comes close to my formulation of a goal for Social Security, though he makes redistribution a goal in itself, for reasons not clear from his discussion.

[11] “The Importance of Social Security Benefits to the Income of the Aged Population,” Irena Dushi, Howard M. Iams, and Brad Trenkamp, Social Security Bulletin, Vol. 77 No. 2, 2017 https://www.ssa.gov/policy/docs/ssb/v77n2/v77n2p1.html

[12] “Policy Basics: Top Ten Facts about Social Security,” Center on Budget and Policy Priorities, March, 2022 https://www.cbpp.org/research/social-security/top-ten-facts-about-social-security.
The OECD defines the poverty level as 50% of the median household disposable income. OECD (2021), Pensions at a Glance: OECD and G20 Indicators, OECD Publishing, Paris, p. 186. By this standard, the incidence of poverty in the United States is much higher across the board, 11.3% for the total population, 19.7% for those aged 65-74, and 28.3% for those 75 and older. Ibid., p. 187.

[13] Ibid.

[14] A Working Paper of the RAND Social and Behavioral Policy Program of the RAND Center for the Study of Aging, November 2020 https://www.rand.org/pubs/working_papers/WRA1009-1.html

[15] Ibid., Abstract.

[16] Steve Goss, Alice Wade, and Jason Schultz, Office of the Chief Actuary, Actuarial Note Number 2004.1, August 2004 https://www.ssa.gov/oact/NOTES/ran1/an2004-1.html

[17] The difference between “unfunded liability” and “transition cost” is in the eye of the beholder. “Unfunded liability” presupposes a liability that should have been funded. “Transition cost” implies moving from pay-as-you-go to a funded system.

[18] Dean R. Leimer, Social Security Administration, Office of Policy, Office of Research, Evaluation, and Statistics, December 2007, p.25.

[19] Cf. footnote 4 above.

[20] A Progressivity Index for Social Security, Andrew G. Biggs, Mark Sarney, and Christopher R. Tamborini, Issue Paper 2009-01, January 2009, p. 4.

[21] Ibid., p. 17. The paper offers much deeper and more specific analysis, and we will return to it when we discuss tax increases required to fund existing benefits. One example: For those currently age 60, the bottom decile is projected to obtain 17% of total benefit, the lower five deciles 59.9% of total benefits, and the bottom eight deciles 85.8% of total benefits.