social security research paper

Strengthening the Social Security safety net

Key takeaways.

• Almost one in every five Americans receives income support from Social Security.
• Social Security is financed on a pay-as-you-go basis, which means that today’s workers’ payroll taxes are used to pay benefits to today’s beneficiaries.
• Changes in population demographics, including longevity improvements, have resulted in a sharp decline in the number of workers per beneficiary , and this trend  is projected to continue.
• As a result, Social Security is facing significant financial challenges and policymakers must take action to ensure its long-term sustainability.

Social Security provides income security to more than 65 million Americans who are either retired or disabled or have experienced the death of a working spouse or parent. The program has reduced poverty among the elderly and given a measure of financial security to those who have paid into the program during the course of their careers. However, Social Security is facing significant financial challenges and policymakers must take action to ensure its long-term sustainability.

This policy brief will address how Social Security works and its importance for retirement.  It will describe how Congress changed the program in 1983 — the last time significant reforms were made — and outline the program’s current financial status as well as the differences between the risks facing Social Security now relative to 40 years ago. The policy brief will conclude with prescriptions and challenges for reforming the program.

Social Security:  The basics

Social Security was created in 1935, began collecting taxes in 1937, and paid its first benefits in 1940. Retirement benefits are calculated based on a worker's highest 35 years of earnings, where past earnings are first indexed upward for the growth of the national average wage. The earnings that determine Social Security contributions and benefits are subject to a taxable maximum, set at $160,200 in 2023. The maximum taxable salary is increased each year along with economy-wide wage growth. Retirement benefits are lowered for those who claim benefits before the full retirement age (67 for those born in 1960 and later), while those who delay claiming benefits receive an increase in their monthly payment.  Since benefits are paid until death, people who choose to claim early receive a lower monthly benefit amount for a longer expected time. 

Social Security was designed to provide a guaranteed income in retirement to protect seniors against the risk of outliving their savings. The program's benefit formula is designed to be progressive and replaces a higher percentage of pre-retirement earnings for lower lifetime earners than for higher lifetime earners.  While Social Security is not an explicit anti-poverty program — in that benefits are not based upon need — the progressive tilt of the Social Security benefit formula has reduced elderly poverty over time. Each year during retirement, the benefits are automatically increased by the rate of inflation.

Social Security is financed primarily on a pay-as-you-go basis, which means that payroll taxes collected from today’s workers are used to fund today's benefits. But Social Security also has two trust funds, the Old Age and Survivors Insurance (OASI) Trust Fund and the Disability Insurance (DI) Trust Fund. The trust funds hold the surplus funds generated by payroll taxes when the program is in surplus and draw down assets to pay benefits when the program runs deficits, as it has since 2010.

Payroll taxes are paid by employees and employers. The OASDI payroll tax rate is 6.2 percent for employees and 6.2 percent for employers for earnings below the taxable maximum ($160,200 in 2023). Self-employed individuals pay both the employee and employer portions of the payroll tax. Social Security also receives income from the taxation of Social Security benefits and interest earned on the trust funds’ assets.

The importance of Social Security in retirement

In 2015, Social Security represented 30 percent of income on average for individuals age 65 and over. Forty percent of all seniors in 2015 received 50 percent or more of their income from Social Security. Meanwhile, 14 percent of all seniors received 90 percent or more of their incomes from the program. [1] Social Security is a critical source of income for those who are widowed and low-income retirees, and Social Security represents a larger share of retirement income for women than for men.

Inequality in life expectancy has significant implications for the distributional consequences of Social Security. Lower-income individuals tend to have lower life expectancies than higher-income individuals, meaning that they may receive fewer years of Social Security benefits. This means that the increasing gap in life expectancy by income works against the progressivity in the benefit formula. [2]   

Figure 1 below shows life expectancy at birth for people at the 10th, 50th, and 90th percentile of the income distribution.  As shown in the figure, life expectancies are higher for those with higher income and, while life expectancies have increased for each group, the gains for those with more income have been larger than for those with less income.

Figure 1 :  Life expectancy by income percentile, 2001-2014

Source: Equality of Opportunity Project

The 1983 amendments and the financial status of Social Security

The 1983 congressional amendments were significant reforms to Social Security, designed to address the program's short- and long-term financial sustainability.

Due to severe inflation and lower-than-expected wages in the early 1980s, the Social Security trust funds were at immediate risk of not being able to pay scheduled benefits in 1983. Insolvency would have prompted short-term reductions to Social Security benefits, though the program was projected to return to solvency by 1990 as the large baby-boom generation reached its peak earning and taxpaying years.

The 1983 reforms addressed short-term solvency by accelerating an increase in the combined employee and employer payroll tax rate from 10.8 to 12.4 percent, previously scheduled to take place by 1990, and by delaying the 1984 Cost of Living Adjustment by six months, a step that amounted to a one-time reduction in current benefits of about 1.75 percent.

Long-term solvency was improved by scheduling a two-year increase in the full retirement age, from 65 to 67, to take place between 2000 and 2022, and making 50 percent of Social Security benefits subject to income taxes for beneficiaries with incomes above $25,000 in nominal terms, a provision that generates increasing revenue over time as nominal income increases raise the share of retirees whose benefits are subject to income taxes.

The 1983 amendments increased revenues for the program, which helped to address the program's long-term funding shortfall, which was a significant concern at the time.  However, by 1984 the Social Security trustees once again projected an actuarial deficit over a 75-year window, with a trust fund depletion date in the mid-2030s projected by the trustees as early as 1993.

The 2023 Trustees Report projected that the OASDI Trust Fund will be depleted in 2034 and that the program faces a long-run actuarial deficit equal to 3.61 percent of wages subject to payroll taxes. This means that the deficit over the 75-year window could be closed either with an immediate and permanent payroll tax rate increase of 3.61 percent; an immediate across-the-board benefit reduction of 21.3 percent; or some combination of current and future tax increases and benefit cuts. The Congressional Budget Office’s Long-Term Projections for Social Security show an even larger deficit of 4.9 percent of taxable payroll. The policy changes would need to be even more drastic in order to balance the program over a longer time horizon, as the shortfall is projected to grow larger over time.

After the trust funds are depleted, Social Security will still be able to pay a portion of promised benefits, with the level of payable benefits decreasing over time as revenues decline. The payable portion of benefits is projected to be 80 percent in 2034, declining to 74 percent by 2097, as shown in Figure 2 below.

Figure 2 :  Old-Age, Survivors, and Disability Insurance (OASDI) Income, Cost, and Expenditures as Percentages of Taxable Payroll

Source:  2023 Social Security Trustees Report, Figure II.D2.

In the 40 years since this last major reform to the Social Security program, people are living longer and fertility rates are declining, resulting in more retirees per worker.  These demographic shifts combined with the pay-as-you-go nature of Social Security — where today’s workers’ payroll taxes are used to fund today's beneficiaries’ benefits — result in a trend toward increasing annual deficits.

Figure 3 :  Fertility, Mortality, and the Number of Workers per Beneficiary

Source:  2023 Social Security Trustees Report, Figures V.A1, V.A5, II.D3.  Total Fertility Rate refers to the average number of children that would be born to a woman if she were to experience, at each age of her life, the birth rate observed in, or assumed for, the selected year, and if she were to survive the entire childbearing period.  Life Expectancy at Age 65 represents average remaining number of years expected prior to death for a person age 65, born on January 1, using the mortality rates for that year over the course of his or her remaining life.  Values beginning in 2022 are projected assuming Social Security’s intermediate cost scenario.

The effect of demographics on Social Security costs can be approximated in the following way. The average Social Security benefit in 2023 is projected to be equal to about 39 percent of the average worker’s wage in that year. If there are 2.7 workers for each beneficiary, as the Social Security actuaries estimate for 2023, the cost per worker is 14.4 percent of their wages (where 14.4 percent represents 39 percent divided by 2.7). If there were five workers per beneficiary, as there were in 1960, that same benefit could be funded at a cost of just 7.8 percent of worker’s wages. Likewise, if the worker to beneficiary ratio falls to 2.3, as is projected by the year 2035, the cost per worker rises to 16.9 percent of wages.

It is important to note that while the retirement of the baby-boom generation accelerated these trends, the underlying changes in mortality and fertility are projected to continue. Thus, Social Security is not expected to return to financial balance even after the baby-boom generation has passed on.

Principles for reforming Social Security

There are several principles that policymakers should consider when evaluating potential Social Security reforms. These include ensuring the program's long-term financial sustainability, maintaining the program's progressivity, protecting vulnerable populations, and avoiding sudden changes that could have significant impacts on beneficiaries.

However, these principles may come into conflict with each other. For instance, phasing in reforms slowly could reduce the impact of such changes on Social Security participants, but would make it more difficult to restore the program to long-run financial health. Likewise, a uniform benefit change that maintained the program’s current level of progressivity could reduce Social Security’s effectiveness at protecting the lowest-income beneficiaries.

Moreover, policymakers may consider more radical changes that alter the broader structure of Social Security. For instance, some countries, such as Australia and New Zealand, fund their retirement programs from general tax revenues rather than from a dedicated payroll tax. Likewise, those countries, as well as the United Kingdom and Canada, focus their retirement expenditures more closely on lower-income retirees while limiting benefits to higher-income seniors.

Barriers to reform and the cost of waiting

There are several barriers to reforming Social Security, including political polarization, the difficulty of making changes to a popular program, and the complexity of the program's finances and benefits.

Delaying Social Security reform could have significant costs. As the population ages, the number of beneficiaries will continue to grow, putting additional strain on the program's finances. Waiting too long to address Social Security's long-term funding shortfall could result in sudden benefit cuts or tax increases, which could have significant impacts on beneficiaries.

In this context, it is worth contrasting the financial situation that will face policymakers when the trust funds are projected to run dry in 2034 versus the conditions that prevailed in 1983, the last time Social Security faced insolvency.

While the 1983 funding crisis was driven by short-term economic changes that drew down the modest trust fund balance then held by the plan, Social Security’s projected insolvency in 2034 is largely the result of demographic changes that have been long in the making. At the same time, the 1983 funding shortfall was substantially more modest than what faces policymakers in 2034.

Had Social Security’s trust fund run out in 1983, Social Security beneficiaries would have faced benefit reductions of about 4 percent on average over a period of six years. By 1990, payroll taxes revenues from the large baby-boom generation would have returned Social Security to annual surplus and the trust funds were projected to remain solvent until between 2025 and 2034.

By contrast, if the trust funds are allowed to become exhausted in the 2030s, beneficiaries would face reductions of around 20 percent in their benefits, roughly five times larger in percentage terms than policymakers faced in 1983. Moreover, those cuts would increase over time, with practically zero possibility that the program would return to solvency on its own without structural changes. Thus, while the 1983 reforms were a model of bipartisan compromise, elected officials face an even larger task in addressing Social Security’s solvency today.

Social Security is an essential program that provides critical support to millions of retirees, survivors, and disabled individuals. However, the program's long-term financial sustainability is at risk due to demographic and economic changes. Policymakers must consider a range of potential reforms to ensure that the program can continue to provide vital support to current and future generations of beneficiaries.

While reforming Social Security is challenging, policymakers must act to address the program's long-term funding shortfall and ensure that the program can continue to meet its important mission.

[1] Dushi, Irena, and Brad Trenkamp. “Improving the measurement of retirement income of the aged population.” Social Security Administration. ORES Working Paper No. 116 (January 2021).

[2] See National Academies of Sciences, Engineering, and Medicine, and Committee on Population. “The growing gap in life expectancy by income: Implications for federal programs and policy responses.” National Academies Press, 2015.

About the Authors

• Gopi Shah Goda is a SIEPR Senior Fellow .  From July 2021 to July 2022, she served as a senior economist at the White House Council of Economic Advisers.  She conducts research that informs how policy can best serve aging populations.  She studies the sustainability of public programs serving the elderly, how individuals make healthcare, saving and retirement decisions as they age, and the broader implications of the COVID-19 pandemic on health, labor supply and entitlement programs. 
• Andrew G. Biggs is a Tad and Dianne Taube Policy Fellow at SIEPR. He is an expert on Social Security reform, state and local government pensions, and public sector pay and benefits. He is a senior fellow at the American Enterprise Institute. In the mid-2000s, Biggs was the principal deputy commissioner of the Social Security Administration, where he oversaw SSA’s policy research efforts.
• SIEPR Research Assistant Bradley Strauss contributed to this policy brief.

Related Topics

• Policy Brief
• Taxes and Public Spending

More Publications

Unequal use of social insurance benefits: the role of employers, the surprising retreat of union britain, the uneasy case for product liability.

Social Security Benefits, Finances, and Policy Options: A Primer

Browse by topic.

• COVID-19 Legislative Response
• Economic Security
• Long-Term Services and Supports
• Medicare and Health Policy
• News About NASI
• Older Workers Retirement Security
• Poverty and Income Assistance
• Social Security
• Unemployment Insurance
• Workers' Compensation
• Workforce Issues and Employee Benefits

Keyword Search

By: National Academy of Social Insurance

Published: September, 2021

**Updated to reflect the estimates of the 2021 Trustees Report .**

The primer is a PowerPoint presentation of approximately 40 slides that provides factual background about Social Security, its benefits and finances, and some policy options to improve the program. The Academy’s income security staff ensures that the data in the presentation are up-to-date.

Topics covered include:

• Who receives Social Security? What are typical Social Security benefits? How do benefits compare to earnings for retirees at different wage levels?
• Who pays for it?
• How many older Americans receive employer-sponsored pensions?
• How are Social Security retirement benefits projected to change in the future?
• What is Social Security disability insurance?
• What are the “best estimate” long-range projections of Social Security finances? What do the high-cost and low-cost projections show? What is the actuarial deficit?
• Why will Social Security cost more in the future? Can we afford Social Security in the future? How can we strengthen Social Security in the future? What are our options? Why consider revenue enhancements to balance Social Security?
• What do American workers say?

Users can download the PowerPoint presentation and sort the slides in a different order or pick and choose a subset to use. The presentation includes talking points that go with each slide. To view the talking points, simply go to View > Notes Page. The sources for the factual material are listed on each slide.

The PDF versions include both the slides and the notes pages that go with them on the same page.

What is Social Security?

The financial outlook, can we afford social security in the future, strengthening social security, briefing: social security financing, benefits, and equity.

This briefing featured an in-depth explanation of the Trustees’ projections of Social Security’s long-term finances. Attendees heard from the Chief Actuary of the Social Security Administration, Stephen C. Goss, and other experts. Discussants answered questions on Social Security’s future financing, benefits, and equity.

Thursday, Sept. 9 | 2:00 – 3:30 ET

How Many People Receive Social Security?

• Retirement insurance
• Survivors insurance
• Disability insurance
• Over 1 in 5 Americans gets Social Security benefits.
• About 1 in 4 families receives income from Social Security.

Social Security Administration (SSA), 2021a; National Academy of Social Insurance, 2020.

Who Receives Social Security?

• 46.9 million retired workers
• 8.0 million disabled workers
• 3.9 million widows and widowers
• 2.3 million spouses
• 1.1 million adults disabled since childhood
• 2.7 million children

SSA, 2021a.

How Much Does Social Security Pay? (December 2020)

* The most recent month-level data presented in SSA’s “Benefits Paid by Type of Family” database as of 9/1/2021 is for December 2020.

SSA, 2021a; SSA, 2021b.

How Do Benefits Compare to Earnings?

SSA, 2021c.

How Many Seniors Rely on Social Security for Most of Their Income?

• Nearly nine out of ten individuals age 65 and older receive Social Security.
• Among elderly beneficiaries, 50% of married couples and 70% of unmarried individuals receive half or more of their income from Social Security.*
• Among elderly beneficiaries, 21% of married couples and 45% of unmarried individuals receive almost all (90% or more) of their income from Social Security.*

*Some evidence indicates that a somewhat lower proportion of beneficiaries, about half, receive half or more of their total income from Social Security, and about 20% may get 90% or more of their income from Social Security (Bee and Mitchell, 2017). This is supported by a 2021 paper which, using IRS data, estimates that about 40% of individuals aged 65 and older receive at least 50% of income from Social Security, 21% receive at least 75% of income, and 14% receive at least 90% of income (Dushi and Trenkamp, 2021).

Reliance on Social Security By Race

*From SSA, on this data:

“In earlier years, 12 of this section’s 20 tables included breakdowns by race. The availability of the race data and the definitions of the race categories varied over time. These concerns about data quality and consistency led to the removal of race breakdowns from this section beginning in the 2011 Annual Statistical Supplement . Although the data quality issues are not yet resolved, Excel versions of those 12 tables are available with the race breakdowns restored. Users are advised to interpret data by race with caution. For more information, see https://www.ssa.gov/policy/docs/ rsnotes /rsn2016-01.html .”

Reliance on Social Security By Gender and Family Type

SSA, 2016: Tables 9.A2 and 9.B3.

Increase in Full Retirement Age (FRA) Lowers Retirement Benefits at Any Age Claimed

Gregory et al., 2010.

Net Social Security Replacement Rates Will Fall

Medium earner’s replacement rate at 65 (after medicare parts b & d premiums and taxation of benefits).

Munnell, 2013.

What is Social Security Disability Insurance?

• Disability Insurance (DI) pays monthly benefits to 8.0 million workers who are no longer able to work due to illness or impairment.
• It is part of the Social Security program.
• Benefits are based on the disabled worker’s past earnings.
• To be eligible, a disabled worker must have worked in jobs covered by Social Security.

What are the Most Common Disabilities for DI Recipients?

SSA, 2020: Table 21

Attributes of Disabled-Worker Beneficiaries

• 3 in 10 disabled workers have incomes below 125% of the poverty threshold.
• older (65% are over 50);
• African-American;
• almost half have a high school diploma or less;
• 10% did not finish high school.

Bailey and Hemmeter, 2015

Who Pays for Social Security?

• Workers and their employers pay with Social Security contributions under the Federal Insurance Contributions Act (FICA).

How Much Do Workers and Employers Pay?

• Workers contribute 6.2% of their earnings for Social Security.
• Employers match these worker contributions (6.2%).
• The total Social Security contribution is 12.4%.
• Earnings above $142,800 are exempt from Social Security contributions.

Board of Trustees, 2021. Table V.C1.

Where Does the Money Go?

• 5.3% goes to the retirement and survivor insurance fund
• 0.9% goes to the disability insurance fund
• Projections of income and outgo of the Trust Funds are made by the Social Security Administration actuaries.

Board of Trustees, 2021: Table II.B2.

2020 Finances

Trust Fund income     =  $1118.1 billion

Trust Fund outgo        =  $1107.2 billion ____________

Increase in Trust

Fund reserves       =    $10.9 billion

• By law, surpluses are invested in U.S. Treasury securities and earn interest that goes to the Trust Funds.

Board of Trustees, 2021: Table IV.A3.

Where is Social Security Income From?

Shares of income to the trust funds, 2020.

What are Social Security Reserves, or Assets?

• Social Security income that is not used immediately to pay benefits and costs is invested in special-issue Treasury securities (or bonds).
• The bonds earn interest that is credited to the trust funds.
• The accumulated surpluses held in Treasury securities are called Social Security reserves, or trust fund assets.
• The Treasury securities are secure investments that are backed by the full faith and credit of the United States government.

Disability Insurance Projections

• Disability Insurance (DI) Trust Fund
• Old-Age and Survivors Insurance (OASI) Trust Fund
• With the Bipartisan Budget Act of 2015, Congress temporarily rebalanced the distribution of Social Security payroll contributions between OASI and DI, extending solvency of the DI Trust Fund.
• According to the 2021 Social Security Trustees Report, the DI Trust Fund is projected to be able to pay full benefits until 2057.

Board of Trustees, 2021: p.4.

How Large are Social Security Trust Fund Assets?

Board of Trustees, 2021: Table IV.A3

Social Security Income and Outgo

How Do Actuaries Estimate the Future?

• Review the past: birth rates, death rates, immigration, employment, wages, inflation, productivity, interest rates.
• Make assumptions for the next 75 years (longer than the rest of the government).
• Three scenarios:
• Intermediate.

Board of Trustees, 2021: Section V.

The Long-Range Intermediate Projection (Trustees’ Best Estimate)

• In 2021, revenue from payroll contributions, interest on reserves, and taxation of benefits is expected to be less than total outgo for the year. If action is not taken in the near future, reserves will start to be drawn down to pay benefits this year.
• In 2034, Trust Fund reserves are projected to be depleted. Income is projected to cover 78% of benefits due then.
• By 2095, assuming no change in taxes, benefits or assumptions, revenue would cover about 74% of benefits due in that year.

Board of Trustees, 2021: Figure II.D2.

Other Scenarios

Low Cost:  The Trust Fund reserves would not be depleted.

High Cost: Trust Fund reserves would be depleted in 2031, instead of 2034.

Board of Trustees, 2021: Table IV.B4.

Why Will Social Security Cost More in the Future?

• Boomers are reaching age 65
• People are living longer after age 65
• Birth rates are projected to remain at replacement levels.
• People 65 and older will increase from 17% to 23% of all Americans by 2095.

Board of Trustees, 2021: Table V.A3.

Percent of the Population Receiving Social Security and Percent Age 65+, 2015-2090

Board of Trustees, 2021: Tables V.A3. and IV.B3.

Can We Afford Social Security in the Future? 

Social security in the broader economy.

Board of Trustees, 2021: Table VI.G4.

Taxable Payroll in the Broader Economy

Board of Trustees, 2021: Tables VI.G4 and VI.G5.

Options to Improve Adequacy

• Updating the special minimum benefit to ensure that long-serving, low-paid workers can remain out of poverty when they retire.
• Reinstating student benefits until age 22 for children of disabled or deceased workers (currently, benefits for these children stop at age 18-19).
• Allowing up to 5 childcare years to count toward benefits.
• Increasing benefits for widowed spouses in low-earning couples.
• Modestly increasing benefits for all by changing the benefit formula (to increase the first PIA bend point by 15 percent)

Options for Raising Revenues

Options that would help raise revenues include:

• Lifting or eliminating the cap (now $142,800) on the earnings on which workers and their employers pay Social Security contributions.
• Gradually increasing the Social Security contribution rate from its current level of 6.2%.
• Subjecting income from investments to Social Security contributions.
• Treating all salary reduction plans like 401(k)s (subjecting income paid into them to Social Security contributions).
• Restoring estate tax to 2000 level and dedicating to Social Security.

Other Options for Solvency

Some proposals would reduce benefits for some or all beneficiaries in order to extend solvency.

• For example, raising the retirement age amounts to an across-the-board cut in benefits, and hence reduces the program’s cost.
• Switching to the chained CPI as the basis for Social Security’s cost-of-living adjustments (COLAs) would reduce benefits and hence program cost as well.

Reno and Lavery, 2009.

Public Opinion on Social Security

Consistent findings throughout the study.

• In 2014, the Academy conducted a multigenerational study to understand Americans’ perspectives on Social Security.
• In focus groups, Americans expressed concern about benefits being too low.
• 77% said it is critical to preserve Social Security benefits, even if it means raising taxes on working Americans.

Walker, Reno, and Bethell, 2014.

In the trade-off analysis, the package preferred by 71% of respondents would:

• for high earners by eliminating the taxable earnings cap;
• For all workers by raising the tax rate by 1/20 of 1% per year.
• For low earners by increasing the special minimum benefit;
• For all beneficiaries by basing COLAs on the inflation experienced by the elderly.

Majorities of Republicans, Democrats, and Independents Agree

Demographic Support for Package of Policy Options Preferred by 71% of Americans

• Benefits are modest (dollars and replacement rates). Yet they are most beneficiaries’ main source of income.
• Social Security benefits will replace a smaller share of earnings in the future than they do today (replacement rates are declining because of the increase in the retirement age) .
• Revenue increases or benefit cuts will be needed to balance Social Security’s future finances.
• Lawmakers have many options to raise revenues, lower future benefits, or increase benefits to improve adequacy.
• Americans value Social Security and are willing to pay for it.
• Americans report they would rather pay more than see future benefits reduced.
• Bailey, Michelle Stegman and Jeffrey Hemmeter. 2015. “Characteristics of Noninstitutionalized DI and SSI Program Participants, 2013 Update.” Research and Statistics Note No. 2015-02. Washington, DC: Social Security Administration. http://www.ssa.gov/policy/docs/rsnotes/rsn2015-02.html
• Bee, Adam and Joshua Mitchell, 2017. “Do Older Americans Have More Income Than We Think?” U.S. Census Bureau Social, Economic, and Housing Statistics Division Working Paper #2017-39, https://www.census.gov/content/dam/Census/library/working-papers/2017/demo/SEHSD-WP2017-39.pdf .
• Board of Trustees. 2021. The 2021 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds . Washington, DC: Social Security Administration.
• Dushi, Irena and Brad Trenkamp, 2021. “Improving the Measurement of Retirement Income in the Aged Population”. Washington, DC: Social Security Administration, Office of Research, Evaluation, and Statistics. ORES Working Paper No. 116. https://www.ssa.gov/policy/docs/workingpapers/wp116.html
• Gregory, Janice M., Thomas N. Bethell, Virginia P. Reno, and Benjamin W. Veghte. 2010. “Strengthening Social Security for the Long Run.” Social Security Brief No. 35. Washington, DC: National Academy of Social Insurance.
• Munnell, Alicia H. 2013. “Social Security’s Real Retirement Age is 70.” Brief No. 13-15. Chesnut Hill, MA: Center for Retirement Research at Boston College.
• National Academy of Social Insurance. 2020. “Social Security Finances: Findings of the 2020 Trustees Report.” Washington, DC: National Academy of Social Insurance. https://www.nasi.org/research/social-security/social-security-finances-findings-of-the-2020-trustees-report-2/
• Reno, Virginia P. and Joni Lavery. 2009. Fixing Social Security: Adequate Benefits, Adequate Financing. Washington, DC: National Academy of Social Insurance.
• Reno, Virginia P., Elisa A. Walker, and Thomas N. Bethell. 2013. “Social Security Disability Insurance: Action Needed to Address Finances.” Social Security Brief No. 41. Washington, DC: National Academy of Social Insurance.
• Social Security Administration. 2021a. “Beneficiary Data: Number of Social Security beneficiaries at the end of July of 2021.” Baltimore, MD: Social Security Administration, Office of the Chief Actuary. https://www.ssa.gov/OACT/ProgData/icp.html
• Social Security Administration. 2021b. “Beneficiary Data: Benefits Paid by Type of Family.” Data for December of 2020. Baltimore, MD: Social Security Administration, Office of the Chief Actuary. www.ssa.gov/OACT/ProgData/famben.html
• Social Security Administration. 2021c. “Replacement Rates For Hypothetical Retired Workers.” Actuarial Note #2021.9. Baltimore, MD: Social Security Administration, Office of the Chief Actuary. https://www.ssa.gov/OACT/NOTES/ran9/index.html
• Social Security Administration. 2021d. “Fact Sheet: Social Security”. Data for December 2020. Baltimore, MD: Social Security Administration, Office of the Chief Actuary. https://www.ssa.gov/news/press/factsheets/basicfact-alt.pdf
• Social Security Administration. 2020. Annual Statistical Report on the Social Security Disability Insurance Program, 2019 . Washington, DC: Social Security Administration, Office of Research, Evaluation, and Statistics. https://www.ssa.gov/policy/docs/statcomps/di_asr/index.html
• Social Security Administration. 2016. Income of the Population 55 or Older, 2014 . Washington, DC: Social Security Administration, Office of Research, Evaluation, and Statistics.
• Walker, Elisa A., Virginia P. Reno, and Thomas N. Bethell. 2014. Americans Make Hard Choices on Social Security: A Survey with Trade-Off Analysis . Washington, DC: National Academy of Social Insurance.

Download this primer.

Download this presentation..

Keywords: Social Security

dddddddddddddddddd

How can we help you?

Stay up-to-date on the latest research & policy updates., subscribe to our newsletter.

Social Security Benefits: An Empirical Study of Expectations and Realizations

I employ data drawn from the Retirement History Survey to study the accuracy of pre-retirement expectations concerning social security benefits. The major findings of this study are as follows. First, survey responses to questions about expected benefits are reasonably noisy. However, when one properly filters out the noise, reported forecasts appear to explain roughly 60% of the variance in realizations. Second, consumers do not form expectations on the basis of all available information. Proper adjustment of forecasts for information contained in concurrent social security entitlements could reduce the residual forecast error variance by roughly 15%. The potential gains from incorporating other information are minimal. Third, individuals do not ignore or forget information which they have used in the past, and they tend to form all expectations on the basis of the same information. Fourth, expectations are highly accurate, given the information that people do use. Extreme optimism is uncommon. Surprisingly, expectations were not abnormally inaccurate during periods of rapid legislative change. Fifth, of various population subgroups, widows and single women tend to make both the most conservative and most accurate forecasts. Married men are least conservative and least accurate. Accuracy and conservativism are not systematically related to wealth or education. Finally, individual behavior appears to conform more closely to the predictions of theory as retirement approaches.

• Acknowledgements and Disclosures

MARC RIS BibTeΧ

Download Citation Data

Published Versions

Issues in Contemporary Retirement, (eds) R. Ricardo, Campbell, E. Lazaer, Hoover Institution Press: Palo Alto, 1988.

Bernheim, B. Douglas and Lawrence Levin. "Social Security And Personal Saving: An Analysis Of Expectations," American Economic Review, 1989, v79(2), 97-102.

More from NBER

In addition to working papers , the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter , the NBER Digest , the Bulletin on Retirement and Disability , the Bulletin on Health , and the Bulletin on Entrepreneurship  — as well as online conference reports , video lectures , and interviews .

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Individuals’ Uncertainty about Future Social Security Benefits and Portfolio Choice

Adeline delavande.

RAND Corporation and Universidade Nova de Lisboa, Economist and Assistant Professor

Susann Rohwedder

RAND Corporation and NETSPAR, Senior Economist, 1776 Main Street, CA 90407, Santa Monica, Tel: 310-393.0411, Fax: 310-451.7084, gro.dnar@rnnasus

Little is known about the degree to which individuals are uncertain about their future Social Security benefits, how this varies within the U.S. population, and whether this uncertainty influences financial decisions related to retirement planning. To illuminate these issues, we present empirical evidence from the Health and Retirement Study Internet Survey and document systematic variation in respondents’ uncertainty about their future Social Security benefits by individual characteristics. We find that respondents with higher levels of uncertainty about future benefits hold a smaller share of their wealth in stocks.

1. Introduction

Planning for retirement involves many intertemporal decisions, most importantly how much money to save and how to invest it. In making these decisions, individuals form expectations about a number of variables, such as future earnings and rates of return. One of these variables is future Social Security benefits, which is the major source of retirement income for 53 percent of married couples and 73 percent of nonmarried persons aged 65 or older ( Social Security Administration, Annual Statistical Supplement, 2009 ). But despite the fact that Social Security benefits are so integral to the economic well-being of older households, previous research has noted high rates of “don’t know” responses from individuals asked to project how much they expect to receive from Social Security. This has raised concerns that people are not well informed about these benefits ( Gustman and Steinmeier, 2004 ).

Lack of knowledge about Social Security rules may explain some of the difficulties people have in forecasting their benefits. But there may be other contributing factors. For example, the level of benefits an individual ultimately collects may depend on events not yet realized before a person first claims Social Security benefits. Uncertainty regarding losing one’s job or experiencing a health shock would likely influence uncertainty about future earnings used to calculate Social Security benefits. Another source of uncertainty may be the possibility of Social Security reform, which both politicians and the media mention frequently. 1 Several reform proposals have been put forward: increasing payroll taxes, decreasing benefits, and introducing personal saving accounts. All of these would have a different influence on levels of benefits.

This uncertainty may have important ramifications, in that it is likely to influence individuals’ decision-making when they plan for retirement, in particular investment decisions. Social Security benefits provide a secure, inflation-indexed, lifetime annuity. If a household could predict with certainty the Social Security benefit amount it will eventually receive, it could adjust the remainder of its investment portfolio accordingly. But what if the household is substantially uncertain about whether and/or how much it can expect? According to a basic prediction of portfolio choice theory—all else equal—someone with little uncertainty about her future Social Security benefits can afford to take on more investment risk in her remaining portfolio than someone who is highly uncertain about his future Social Security benefits. 2 But because data are few, little is currently known about the extent of the uncertainty, how it varies across individuals, and how it influences economic behavior.

In this paper, we use newly available data eliciting individuals’ subjective distributions about their future Social Security benefits first to describe their uncertainty about their future Social Security benefits and how it varies with observable characteristics, and second, to investigate whether the uncertainty appears to be related to households’ portfolio choice. More specifically, we assess whether there is support for the hypothesis that individuals with higher uncertainty about future Social Security benefits tend to invest a smaller fraction of their wealth in risky assets. The data come from a 2007 supplemental Internet survey of a subsample of respondents to the Health and Retirement Study (HRS). The HRS is a large longitudinal survey of the U.S. population over the age of 50. We designed the module on Social Security expectations in the 2007 Internet survey, employing an innovative visual format to elicit the distribution of individuals’ subjective beliefs about the benefit amounts they might receive.

Eliciting the whole subjective distribution of future Social Security benefits has rarely been done. Several large-scale surveys—such as the core survey of the Health and Retirement Study, the Survey of Consumer Finances, and the Retirement History Survey—have asked individuals to forecast what they expect to receive in future Social Security payments. But the question design in these surveys leaves no room for expressing uncertainty, instead querying only about point estimates (e.g., “ How much do you expect your monthly Social Security benefits to be? ”). These data on point estimates have been used to investigate whether Social Security depresses savings (e.g., Bernheim and Levin, 1989 ; and Bottazzi et al., 2006 ) and whether it affects portfolio choice ( Bottazzi et al., forthcoming ).

A notable exception, however, is the Survey of Economic Expectations (SEE). Dominitz and Manski (2006) , who designed the module on Social Security expectations in the SEE, elicited the minimum and the maximum and six points on respondents’ subjective probability distribution of their future Social Security benefits. Based on this information they derived measures of uncertainty for each respondent. They found that both younger and older respondents showed substantial uncertainty about future Social Security benefits. They also noted that younger individuals appeared to be concerned that the Social Security system might not survive overall, but not that benefits would be reduced should the system survive.

This paper complements, as well as extends, the findings reported in Dominitz and Manski (2006) in two ways: First, it correlates individuals’ uncertainty about their Social Security benefits with a rich set of covariates from the HRS, such as the subjective probability of a Social Security reform. Second, it studies the relationship between the measures of uncertainty and economic decision-making, using the example of portfolio choice. We are aware of only one other study that relates uncertainty about future public pension benefits to economic decision-making. Guiso et al. (2009) analyze the subjective distribution of public pension replacement rates among Italian workers using data obtained in a survey of a random sample of customers of a large financial institution. 3 They find that higher pension risk is associated with increased investments in life insurance and with a larger likelihood to enroll in private pension funds and private health insurance plans.

The paper is organized as follows: In the next Section, we present a theoretical model of portfolio choice in which retirement age is endogenously determined. While simple, it allows us to derive our main hypothesis of how uncertainty about Social Security benefits might influence the fraction of wealth invested in risky assets. It also provides guidance regarding other factors that should enter the empirical analysis. In Section 3, we then introduce the data, discuss our analytical sample, and describe the innovative visual design we used to elicit Social Security expectations. We provide detailed descriptive statistics of how the subjective probability of receiving any Social Security benefits in the future, and individuals’ uncertainty about the expected benefit amounts, both vary with observable characteristics in a multivariate setting. Note that when eliciting the uncertainty about the expected benefit amount, one could ask respondents to report the amount they expect to receive either conditional on starting to receive these benefits at a particular age or without this conditioning. In the HRS Internet module, we asked both versions of the question, randomizing across respondents. To show to what extent this leads to different empirical patterns, we present results for both.

In Section 4, we present predictors of uncertainty about future Social Security benefits. Then in Section 5, we turn to our analysis of portfolio choice and how it relates to uncertainty about expected Social Security benefits. We estimate reduced-form models that incorporate the basic implications derived from the theoretical framework. More specifically, we relate the fraction of total (non-pension) wealth held in stocks to the respondent’s uncertainty about future Social Security benefits and other covariates. A challenge in the econometric analysis is that the uncertainty in Social Security benefits may be driven by uncertainty about when to claim, which may be endogenous. To deal with this endogeneity issue, we take advantage of the fact that we have various elicitation designs and mitigate the endogeneity issue by focusing on respondents who were asked their distribution of Social Security benefits conditional on a particular claiming age. Finally, in the Conclusions, we summarize our findings and suggest future avenues of research.

2. A Simple Model of Portfolio Choice

It is challenging to model how households allocate their investment portfolios, because some households seem to make investment decisions that are difficult to reconcile with existing theories—for example, choosing not to participate in the stock market (see Campbell, 2006 , for an overview of the literature). We do not aim to develop a complete model of portfolio choice. Rather, we seek to illuminate how a household’s uncertainty about future pensions or Social Security benefits may influence its decisions about investing in risky assets. Because we focus on Social Security benefits, and because Social Security benefits depend on the timing of retirement, we relax the assumption of fixed labor supply common in models of portfolio choice in the finance literature (see Merton, 1990 , for an overview). This is to allow for the possibility that there may be a relationship between portfolio choice and labor supply, in particular the timing of retirement: Working longer may be a form of insurance against the poor performance of previous investment decisions. Bodie and Samuelson (1989) and Bodie, Merton, and Samuelson (1992) develop theoretical models showing that for individuals with more labor supply flexibility, it is optimal to invest a larger fraction of their wealth in risky assets.

We present a simple two-period model explaining individuals’ choices regarding their portfolio allocation, consumption, and retirement age. Our model is based on the two-period model developed by Bodie and Samuelson (1989) . In period 1, the individual allocates his current wealth between a risky and a risk-free asset. At the end of period 1, the rate of return of the investment is realized. Given the resulting wealth, the individual decides how much to consume and when to retire. Accordingly, two important aspects of the model are that the timing of retirement is flexible, and that it is determined after the uncertainty about investment returns has been resolved. For simplicity, the individual is also assumed to claim Social Security benefits when he retires, and to consume all of his wealth in the second period.

This individual is endowed with wealth W 0 at the beginning of period 1, and allocates a share 1 − x of his wealth to a risk-free asset with a rate of return r , and a share x of his wealth to a risky asset yielding z 1 with subjective probability p 1 and yielding z 2 with subjective probability p 2 = 1 − p 1 . We assume that the subjective expected return of the risky asset exceeds that of the risk-free asset: p 1 z 1 + p 2 z 2 > r and that z 1 < r < z 2 . In period 2, the individual can decide to retire at any age A that we normalize so that it is between zero and 1. While still working between the ages 0 and A in period 2, he earns the wage Aw . After retiring, he earns Social Security benefits and enjoys leisure L = 1 − A in retirement. The amount of Social Security he earns depends on how long the individual has been working. For simplicity, we assume that the Social Security formula is linear in A so that an individual who retires at age A gets a lifetime benefit equal to B × A , where 0 < B < w . 4

We first consider the case in which the individual does not face any uncertainty about future Social Security benefits. The individual maximizes his expected utility. His utility function U ( C, L ) depends on consumption C and leisure L . We assume that U is increasing, concave, and differentiable in both arguments. The individual solves the following problem:

The first order conditions with respect to L 1 , L 2 and x are:

The first-order condition with respect to x implies that the risk-averse individual will invest a positive share in the risky asset. This is because at x =0, we have C 1 = C 2 and as a result, L 1 = L 2 . So V x > 0 at x =0. For the first-order condition to hold, one needs to increase x .

We next consider the case in which there is uncertainty in Social Security benefits. Suppose that Social Security benefit b is a random variable that is independent of z and has the same expected value as in the certainty case above E ( b ) = B . The random variables b and z are assumed to be realized at the same time. We have the following results:

Proposition 1

Suppose the individual’s utility function is of the following form:

with α > 0 , β > 0 , 0 ≤ C 0 < W 0 , 0 ≤ L 0 < 1 − ( 1 + z 2 B + w ) .

Introducing uncertainty about Social Security benefits reduces the optimal fraction of total wealth invested in the risky asset.

The proof is presented in Appendix A .

Note that the utility functions specified above comprise utility functions that have constant relative risk aversion with respect to C when C 0 = 0, and utility functions that have decreasing relative risk aversion with respect to C when C 0 > 0. 5

This simple model implies that higher uncertainty about future Social Security benefits should be associated with a smaller fraction of total wealth invested in stocks. In Section 5, we will assess whether there is empirical support for this. For simplicity, we assume that there is no correlation between stock market returns and Social Security benefits. A positive correlation between the two should reduce the individual’s demand for the risky asset.

As pointed out by Bodie and Samuelson (1989) , this model further implies that younger workers can take on more investment risk than older workers, because they have a longer time horizon during which they can adjust their labor supply in response to any adverse portfolio outcomes.

It is also useful to establish in the context of this model what quantities the individual would report if asked in period 1 for his or her subjective expectations about Social Security benefits. The benefit amount depends on the retirement age which in turn depends on the investment returns which are not yet known in period 1. Therefore, the individual would report the distribution of p 1 bA 1 + p 2 bA 2 . Note that it depends on the individual’s expectations about stock market returns in period 1, given by p 1 and p 2 . However, if the individual was asked about his or her subjective distribution of Social Security benefits conditional on claiming at age A 1 , the individual would report the distribution of bA 1 . The distribution of benefits conditional on a claiming age has therefore a smaller variance than that of the distribution unconditional on a claiming age, and does not depend on expectations about stock market returns.

3. Data on Social Security Expectations

3.1. the hrs internet survey and our analytical sample.

We use data on Social Security expectations that come from the second wave of the HRS Internet Survey, a supplementary survey of the Health and Retirement Study (HRS). The HRS is a panel survey representative of the U.S. population age 51 and over. In the core survey, the HRS collects biennial data on close to 20,000 individuals and their spouses in about 13,000 households. 6 Eligibility for the second wave of the HRS Internet Survey was determined by whether a respondent reported in the HRS 2004 or 2006 core survey that he or she regularly used the Internet. Eligible respondents were divided into two groups. The first was invited to participate in the survey in the spring of 2006 (called phase I) and the second in the summer of 2007 (called phase II). 7 We link the data from the HRS Internet survey to the HRS 2006 core survey to obtain additional covariates on the same respondents. Therefore, our various samples of analysis are composed of those respondents who answered the relevant questions about Social Security in the HRS Internet survey and who also responded to the HRS 2006 wave.

In this paper, we focus on two aspects of respondents’ Social Security expectations: (i) the subjective probability of receiving Social Security benefits at some time in the future and (ii) the subjective distribution of future Social Security benefits, conditional upon receiving them. Only respondents who were not currently receiving Social Security benefits were asked about these items. Exactly 1,500 respondents from phases I and II of the second wave of the HRS Internet survey reported not receiving Social Security benefits at the time of the interview. 1,489 provided a percent chance when asked about the probability of receiving Social Security benefits in the future. Of those, 5% said there was a zero percent chance that they would receive Social Security benefits in the future. Consequently, this 5% were not asked any further questions about their expectations regarding Social Security benefits.

For the analysis of uncertainty about future Social Security benefit amounts our analytical sample is smaller. This is because we experimented with two alternative formats for their elicitation. Half the sample was administered a sequence of percent chance questions similar to the design in the SEE by Dominitz and Manski, and the other half was administered a visual format that we designed and describe in detail below. 8 In this study we focus our analysis of uncertainty about future Social Security benefits on respondents who were randomized into the visual format (783 respondents, of whom 774 also participated in the HRS 2006 core survey). As noted in Section 2, one can ask individuals about the benefit amount they expect to receive with or without conditioning on a particular age at which the individual would first receive the benefits. We did both and once again randomized which set of respondents received which format (nobody was asked both). We take advantage of the availability of observations of uncertainty both conditional and unconditional upon the expected claiming age (albeit on different samples of respondents) in our empirical analysis of portfolio choice and how it relates to uncertainty about future Social Security benefits. Item non-response on the Social Security-related questions in the HRS Internet survey is very low (less than 2%).

Table 1 gives the characteristics of our analytical sample. 1,479 of HRS Internet respondents report a probability of receiving Social Security in the future and answered the 2006 HRS core survey. 9 Among those, the average and median age is 56 (90% of our sample is between 50 and 65), 62% are female, 78% are married, 84% are working for pay (and the average length of working life to date is 29 years). The number of years spent in education is 14 years both at the mean and at the median. When looking at the highest achieved degee, 27% have a high-school degree or less, 32% have some college, and 41% have completed college or more. Characteristics of those randomized into the visual format are similar.

Descriptive statistics

The number of observations varies across variables ranging between 313 and 1,479. We present the statistics for each variable over the sample for which it is available combined with any other sample restrictions applicable to the subsequent analyses where this variable appears. For example, for Social Security benefit expectations conditional on claiming age the number of observations is N=316; for the chance of receiving Social Security in the future the number of observations is N=1,479.

As one would expect for a sample selected on the basis of Internet usage, this pool of respondents is more educated and more likely to work than the general population of the same age. For example, among the HRS 2006 respondents between the ages of 50 and 65 who do not currently receive Social Security, 39% have a high-school degree or less, 28% have some college, 32% have completed college or more, and 79% are currently working. 10

3.2. Expectations about Social Security: Survey Design and Descriptive Statistics

All HRS Internet respondents in phase I and phase II who reported not currently receiving Social Security benefits were asked:

On a scale from 0 to 100, (where 0 means no chance and 100 means absolutely certain), what do you think is the percent chance that you will receive Social Security benefits some time in the future?

Most respondents gave a high chance of receiving Social Security benefits in the future. The median and average subjective probabilities are 90% and 79% respectively ( Table 1 ). Five percent of respondents answered 0%, 13 percent said 50%, and 40 percent said 100%.

Respondents who provided a positive probability of receiving Social Security at some time in the future were then instructed to assume for subsequent questions that they would, in fact, receive benefits:

Suppose that you will indeed receive SS benefits in the future. We will ask some questions about when you expect to receive them and how much you think they will be.

These respondents were then asked about the timing of claiming:

At what age do you expect to start collecting these benefits?

In phase II of the HRS Internet survey, we used a visual format to elicit information on individuals’ subjective probability distributions about their future Social Security claiming age in a way that mimicked the density function of their subjective beliefs. 11 We asked respondents to allocate 20 balls into 8 bins, each representing a claiming age, to express the likelihood that they would first claim benefits at each age. The left outer bin represented age 61 or younger, while the right outer bin represented age 68 or older. On a first screen, we presented a short introduction to familiarize respondents with the exercise. The next screen, replicated in Figure 1 , showed an example with 12 balls allocated in the bin for age 64, and 8 balls in the bin for age 65, and described this particular allocation. Finally, on the third screen—similar to that of Figure 1 , except with all 20 balls placed in a box beneath the bins—respondents were asked to allocate all the balls into the 8 bins to express the chances out of 20 that they would claim at each age.

The distribution of point estimates for the claiming age exhibits heaping at ages 62 (27% of answers) and 65 (30% of answers). Only 14% report their normal retirement age at which they qualify for full Social Security benefits as their expected claiming age. Respondents tend to have subjective probability distributions concentrated around the reported expected claiming age: 47% of respondents allocate all of the probability mass within one year of the reported expected claiming age and 76% allocate at least 75% of the probability mass within one year of the expected claiming age.

Next we asked respondents about the amount of their expected future Social Security benefit. The elicitation method followed a sequence similar to that in the elicitation of the distribution of the expected claiming age. Respondents who reported a positive probability of receiving Social Security benefits in the future were first asked to give a point estimate about their future benefits, conditional on receiving any:

How much do you expect your monthly Social Security benefits to be in today’s dollars?

For half the sample we then again used a visual format to elicit the subjective distributions of benefits conditional on future receipt. 12 Respondents were presented with an introduction and an example and were asked to allocate 20 balls into the 7 bins so that the allocation would reflect the chances out of 20 that their monthly Social Security benefits would fall into any one of the possible dollar ranges (see Figure 2 ). The example was the same for all respondents. But on the screen where respondents would allocate the balls themselves the thresholds for the bins were tailored to each individual respondent and centered around his or her point estimate of expected future Social Security benefits. The left outer bin ran from zero to 30% of the respondent’s point estimate (rounded to the next 5 dollars); each of the middle five bins covered an equal range (interval amounting to 30% of the respondent’s point estimate). The right outer bin was open-ended, as shown in Figure 2 . If a respondent did not provide a point estimate, we administered standardized thresholds centered around $1,100.

Phase II respondents who allocated all of the balls into one or two adjacent bins were presented with a follow-up bins-and-balls screen. These “unfolding bins” split the initial range containing all the balls into narrower bins of equal width. The respondents were asked to place 20 balls into those narrower bins (see Delavande and Rohwedder, 2008 , for the exact wording presented to the respondents). The number of new bins varied between 2 and 5, depending on the size of the range to which the respondent had initially allocated all of the probability mass. 13

The design above elicits expected Social Security benefits without asking respondents to assume that they would claim at a particular age. Within the sample of respondents assigned to the visual format, respondents were further randomized into two subgroups: 59% of the sample were asked to report their distribution of Social Security benefits unconditional on their claiming age, as just described above; the other 41% were asked to report their distribution of future Social Security benefits conditional on claiming at their point estimate of their expected claiming age. 14 We conducted this experiment because little is known about how conditioning on a particular claiming age affects the distribution of Social Security benefits elicited. 15 The timing of when a person first claims Social Security benefits determines the benefit amount in important ways—first, because of how benefits are computed and second, because of the potential additional earnings for respondents who work longer. For someone who first claims at the normal retirement age (NRA, which is 66 for most respondents in our sample), 16 the monthly benefit equals 100% of the Primary Insurance Amount (PIA), which itself depends on past earnings. However, individuals can claim benefits as early as age 62. 17 Holding retirement age and lifetime earnings constant, if an individual claims between age 62 and his or her NRA, there is an actuarial reduction in the benefit for each month of claiming before the NRA. 18 If he or she claims after the NRA, there is a delayed-retirement credit that increases until age 70. Most people start claiming Social Security benefits at about the same time as they retire. As a result, the PIA will differ if someone claims early rather than late, due to the additional earnings. A person’s uncertainty about his or her expected Social Security claiming age therefore ought to be strongly reflected in that person’s uncertainty about the benefit amount unconditional on claiming age.

The 25 th , 50 th , and 75 th percentiles of the distribution of the expected monthly benefit amount unconditional on the expected claiming age are $750, $1,100, and $1,500 respectively. For the distribution of the expected monthly benefit amount conditional on expected claiming age, those same percentiles are closely comparable: $750, $1,000, and $1,500 ( Table 1 ). Both distributions—conditional or unconditional on expected claiming age—tend to center around the point estimate. 19 Respondents allocate, on average, 60.3% of the probability mass to the middle bin (which covers the range of plus/minus 15% around the point estimate), 17.2% to the bin immediately to the left of the middle bin, and 16.6% to the bin immediately to the right of the middle bin. The outer bins attract low probability mass overall.

4. Predictors of Uncertainty about Social Security Benefits

Intuitively one can think of a number of factors that might affect individuals’ expectations about Social Security: for example, outcomes of future events that affect the calculation of benefits (e.g., future earnings or the timing of retirement), the possibility of a recession affecting earnings and therewith Social Security benefits; or political risk with respect to Social Security reform. But there could also be variation in expectations of a very different nature resulting from inattention. For example, a very wealthy person might not care to know what amount he or she is entitled to because it accounts for too small a fraction of retirement resources. Research on models of expectation formation is still in its infancy. Therefore we do not know exactly how these factors might enter the expectation formation process. In what follows, we show descriptive multivariate regressions of Social Security expectations on potential determinants.

The uncertainty that so many respondents show about their future Social Security benefits can be decomposed into their uncertainty about eligibility and their uncertainty about the benefit amount conditional on being eligible. Because each of these two components is of separate interest we conducted separate analyses of how they relate to observable characteristics. For completeness, we also analyzed how characteristics are related to the central tendency of the distribution of Social Security benefits.

We used the elicited subjective probability of receiving Social Security benefits in the future as a measure of uncertainty about eligibility. To facilitate our analysis of the distribution of benefits, we made parametric assumptions similar to those of Engelberg et al. (2009) to fit a distribution to respondent’s answers. In particular, we assumed that the subjective distributions with positive probability in at least three bins were of generalized beta form, and that the subjective distributions based on one or two bins were isosceles triangular (See Appendix B ). We then used the mean of the fitted distribution as a measure of central tendency, and the standard deviation as a measure of uncertainty. If we use the median and interquantile range instead, we find similar results (results not shown).

Regarding the uncertainty about Social Security benefits, we conducted separate analyses for respondents who were asked their distribution conditional or unconditional on their expected claiming age. When we asked respondents to report their distribution conditional on a particular claiming age, they presumably provided a distribution of beliefs that does not incorporate uncertainty about the claiming age.

In this context, a question arises: Which feature of the distribution of claiming ages (e.g., the mean, mode, median, or another quantile) do respondents report when they are asked what they expect the point estimate of their claiming age to be? We investigated this question by assuming that the distribution on claiming age is discrete and that its support is between 60 and 70 (see Appendix C for the details of these assumptions). We find that the reported point estimate for the claiming age is consistent with the mode for 90% of the respondents, with the median for 76% of the respondents, and with the mean for 52% of the respondents.

Table 1 shows some descriptive statistics of the standard deviations of the fitted distributions. Several patterns are of interest. First, respondents exhibit considerable uncertainty about their future monthly benefit amount conditional on being eligible: The average standard deviation is about $105, which is about 9% of average expected benefits. 20 Second, the degree of uncertainty varies within the population: The interquantile range of the standard deviation ranges between $83 and $88. Finally, all of the quantiles of the distribution of standard deviations are higher for the unconditional format than for the conditional format, suggesting, as one would expect, that respondents who are asked their beliefs unconditional on a claiming age exhibit more uncertainty.

Table 2 presents the Best Linear Predictors under square loss that relate the various measures of Social Security expectations to a number of variables: socio-economic characteristics; factors that influence eligibility according to current rules, such as the total number of years worked; future earnings, such as the subjective probability of losing one’s job or the probability of an economic recession in the next 10 years; and the subjective probability that Social Security reform will take place in the next 10 years so that benefits become less generous than they are now. Table 1 provides descriptive statistics of those variables.

Best Linear Predictors under Square Loss of Social Security Expectations

Robust standard errors in brackets. Regressions include indicators for missing variables.

ECA = expected claiming age; SS = Social Security.

The first column in Table 2 shows that age is an important predictor of the probability of receiving Social Security in the future. Respondents in their late 50s and early 60s provide a higher probability of receiving future benefits than those in their early 50s or younger. Respondents age 50 to 54 have, on average, a subjective probability of benefit receipt 6 percentage points lower than that of respondents age 55 to 59. This is consistent with the fact that an individual’s uncertainty about what the structure of Social Security will look like at the time of his or her first claim should decrease as the time of claiming draws near.

In addition, variables related to current eligibility rules are clearly important predictors: Respondents who work, and those who have been working longer, report a higher probability that they will receive Social Security in the future. On average, respondents who work provide a subjective probability of eligibility 8 percentage points higher than those who do not work. This is consistent with the fact that under current law, only individuals who have worked 10 years or more are eligible to receive benefits based on their own work record. Concerns about Social Security reform, as well as about the state of the economy (which may influence the type of reform that takes place), are associated with the subjective probability of being eligible. Conditioning on other covariates, those who think there is a higher chance of Social Security reform and of a recession within the next 10 years provide a lower probability that they will receive any benefits in the future.

The second column of Table 2 shows how observable characteristics relate to the mean of the fitted distribution of the benefit amount. For this variable, we pooled respondents from the conditional and unconditional format together, because the format does not seem to influence how characteristics are related to the mean (tables not shown). We did, however, add an indicator for the elicitation format. Variables that are known to affect lifetime earnings (and thus the PIA), such as gender, education, and number of years worked, are dominant predictors. In addition, individuals who report a lower subjective probability of Social Security reform have a lower mean. This finding, combined with the results shown in Table 1 , is consistent with the idea that individuals fear that the system will either disappear entirely, or that benefits will be reduced.

Looking at the last two columns of Table 2 , we see that the important predictors of the probability of eligibility are different from those of the uncertainty about the amount conditional on being eligible. Interestingly, the coefficients associated with the subjective probability of an economic recession and Social Security reform are not precisely estimated in the benefit amount specification. But this result may be due to the smaller sample size.

An important predictor of uncertainty about the amount of future benefits is the individual’s distance to the expected claiming age. 21 The coefficients of Table 2 are consistent with the idea that those who are further away from claiming face more unresolved uncertainty about events that may influence the benefit amount. But even people who are less than 5 years away from retirement still report uncertainty (in this group, the average standard deviation of monthly benefits is $90). In general, women—in addition to reporting a lower average benefit amount than men—report less uncertainty about future benefit amounts than their male counterparts.

The coefficients in columns 3 and 4 exhibit similar patterns overall. In Column 4, we see that the uncertainty about the timing of claiming, as measured by the standard deviation of the distribution of claiming ages, is positively associated with uncertainty about the expected benefit amount and has a very large coefficient. 22

Lack of knowledge of the rules that govern Social Security benefits may be an important determinant of uncertainty. But unfortunately, we have limited data on people’s knowledge about the Social Security program. We observe whether a person knows that filing claims early results in reduced benefits: 23 We have created an indicator for knowledge about this feature of the rule. It turned out not to be a statistically significant predictor of uncertainty and including it did not change any other coefficients. We therefore do not present it in Table 2 .

5. Uncertainty about Future Social Security Benefits and Portfolio Choice: Empirical Analysis

Social Security benefits are the main source of retirement income for the majority of retirees. But in Section 4, we show that many respondents—including those close to their expected claiming age—exhibit uncertainty about the level of benefits they expect to receive. The theoretical model we presented in Section 2 indicates that individuals who face more uncertainty in their Social Security benefits invest a smaller fraction of their wealth in risky assets. In this section, we investigate this premise empirically. To this end, we estimated a reduced-form model, taking into consideration the uncertainty that individuals face with respect to both the uncertainty related to their eligibility for benefits and that related to the benefit amount, i.e., the distribution of benefits unconditional on eligibility. 24 We estimated an equation of the form:

where x i denotes the fraction of wealth held in stocks by the household in which individual i lives, sd i denotes the standard deviation of i ’s subjective distribution of Social Security benefits, Z i denotes individual and household characteristics, and ε i is an unobserved random term. On average, individuals in our sample hold 14% of their total assets in stocks; the median is 3% ( Table 1 ). Forty-five percent of the households in which our respondents live do not hold any stocks. (Note that the fraction of wealth invested in stocks is a household-level variable from the HRS 2006 core survey). 25 In addition to uncertainty about their own future benefits, married individuals in our sample may have uncertainty about their spouse’s Social Security benefits. However, we do not have any information on this variable.

As we explained in Section 3.2, information is available on the distribution of Social Security unconditional on the expected claiming age (i.e., the distribution of p 1 bA 1 + p 2 bA 2 in the model) for some respondents, and for other respondents, conditional on an expected claiming age (the distribution of bA 1 , for example, if the respondent reports the most likely claiming age as expected claiming age and p 1 > p 2 ). In this simple framework, answers about bA 1 should capture the uncertainty associated with b . In practice, an individual who delays claiming may work longer. As a result of the additional years of earnings, his PIA may increase. But because there may be uncertainty about future earnings close to retirement (i.e., in w in our model), this individual may exhibit more uncertainty about future Social Security benefits conditional on claiming late than conditional on claiming early. As such, the standard deviation of the distribution unconditional on a claiming age—which supposedly also captures uncertainty about lifetime earnings for all potential claiming ages—would better capture the uncertainty that individuals consider when making their portfolio choice than would the standard deviation of the distribution conditional on a claiming age.

However, using the standard deviation of the distribution unconditional on the expected claiming age poses some econometric challenges. By definition, this standard deviation depends on the subjective distribution of stock market returns, because those influence the retirement age, which is not precisely measured in the HRS. As a result, it may be correlated with the random term ε i . To deal with this issue, one could use an instrumental variable approach, but we do not have a suitable instrument. Instead, we used the standard deviation of the distribution of Social Security benefits conditional on a claiming age. The underlying assumption is that this uncertainty, which depends on factors such as the probability of reform, is exogenous to unobservable variables that influence portfolio choice. But this standard deviation may underestimate the overall uncertainty that individuals face. Consequently our estimate may provide a lower bound of the effect of uncertainty about future Social Security benefits on portfolio choice.

Our theoretical framework suggests a number of other covariates we should include in our reduced-form model. People’s subjective beliefs about stock market returns are critical to their portfolio choice. We used (1) the elicited subjective probability that the price of mutual fund shares invested in blue chip stocks, such as those in the Dow Jones Industrial Average, will increase faster than the cost of living over the next 10 years, (2) the subjective probability that it will increase by 8 percent or more per year on average over the next 10 years, and to complement these measures (3) the subjective probability that there will be an economic recession within the next 10 years.

Risk-aversion is another important factor in portfolio choice and we used it too as a covariate. The HRS includes a categorical measure of risk-aversion derived from a set of questions where the respondent is asked to choose between two jobs, where one guarantees the current family income and the other offers a chance to increase income, but also carries the risk of income loss. As we point out in Section 2, the time horizon may additionally be important, so we controlled for age and a measure of subjective survival expectations relative to the life table (i.e., the ratio of the subjective probability of being alive at age 75 to the probability given by the life table). We used education as a covariate as well, because a more educated individual may be more able to insure his risky investment with labor income. Finally, we controlled for the level of total wealth, including expected Social Security entitlements. We created an indicator for total bequeathable wealth (above and below the median in our sample, referred to as “other wealth” in Table 3 ) and an indicator for Social Security wealth as measured by the mean of the fitted distribution of Social Security benefits (above and below the median), and interacted the two. We also included an indicator for whether the individual has an employer pension.

Best Linear Predictors under Square Loss of the Fraction of Wealth held in Stocks

Robust standard errors in brackets.

Regressions include indicators for missing variables.

Table 3 shows the best linear predictors of the fraction of wealth held in stocks under square loss. The first column uses the standard deviation of the fitted distribution of Social Security benefits unconditional on claiming age as the measure of uncertainty, while the second column uses the standard deviation of the distribution conditional on the expected claiming age. The second column therefore presents our preferred estimates, as conditioning on the expected claiming age reduces the problem of endogeneity. Focusing on this column, we find, as predicted by the theoretical model, that individuals with more uncertainty about their future Social Security benefits are less likely to hold a greater portion of their wealth in stocks, and the coefficient is statistically significant at 5%. All else equal, increasing the standard deviation from the 25 th to the 75 th percentile reduces the fraction of wealth held in stocks or bonds by 0.017. 26

Turning to those expectations related to the stock market or economy, we see that individuals who report a higher subjective probability that the price of stocks will rise faster than the cost of living over the next 10 years, or a lower subjective probability of a recession, hold a higher fraction of their wealth in stocks. Contrary to what the model predicts, Table 3 shows that older individuals have a higher share of their assets in stocks. This may be due to the fact that younger households may be more likely to face borrowing constraints, which may make risky financial investments less attractive to them ( Campbell, 2006 ). However, individuals with more education tend to hold a higher proportion of their assets in stocks. The coefficients associated with the measures of risk aversion here are not precisely estimated, which may be due to the fact that the hypothetical questions about job loss do not do a very good job of capturing risk-aversion related to financial investments.

Table 3 also shows interesting patterns related to wealth: Those with high other (bequeathable) wealth and high Social Security wealth invest by far the largest proportion of their assets in stocks (almost 10 percentage points more than the other groups).

The coefficients associated with the standard deviations of the distribution of benefits unconditional on claiming age is much smaller in magnitude than the one for the distribution conditional on claiming age. This may suggest that endogeneity is indeed an issue with the unconditional distribution, and that using the distribution conditional on an expected claiming age mitigates that problem.

Overall, the results presented in Column 2 based on the conditional distribution favor the hypothesis that individuals who exhibit less uncertainty about their future Social Security benefits hold more risky investment portfolios. Under the assumption that using the conditional distribution eliminates the endogeneity problem and that no other omitted variables bias the estimated relationship, these relationships can be interpreted as causal. In that sense, our results suggest that our respondents act qualitatively as theory would predict. However, they do not allow us to determine whether individuals choose the optimal amount of risky investment given the Social Security benefits they expect.

It is possible, though, that these results could be driven by some omitted variables. This would happen if, for example, lower levels of uncertainty were positively correlated with higher financial literacy or cognitive ability (and thus better knowledge of Social Security rules), and, in turn, if individuals who are more financially savvy are more likely to hold a larger portion of their wealth in stocks. Our specification includes controls for gender and education, which have been found to be important indicators of financial literacy ( Lusardi and Mitchell, 2007 ). 27 But we cannot rule out that this may be an important variable whose effect is not fully captured by the controls for gender and education.

Another issue may be related to the correlation between expectations about stock market returns and Social Security benefits. Such a correlation would occur, for example, if people believed that a correlation exists between the growth of their earnings and returns on stocks, or that a Social Security reform that would reduce benefits is more likely to happen when stock returns are low. In Table 1 , we find that individuals who report a higher probability of a recession within the next 10 years report a lower probability of eligibility, which may indicate that a positive correlation does exist between stock market returns and Social Security expectations. This correlation would be problematic for our analysis of portfolio choice if it were to vary systematically across individuals and influenced their portfolio choice. But at this point, we do not have the information on the joint subjective distribution of benefit amounts and stock market returns that would enable us to establish whether this is indeed the case.

6. Conclusions

In this paper, we have presented empirical evidence concerning the uncertainty individuals have about their future Social Security benefits, and how this uncertainty influences their portfolio choice. The evidence is based on a new data collection method that we applied in the HRS Internet survey.

We document sizeable uncertainty about respondents’ eligibility for Social Security benefits, as well as the amount of benefits they expect to receive. We find that younger respondents and those further away from their expected claiming age exhibit greater uncertainty, suggesting that uncertainty related to events that have not yet happened may be relevant. In line with this idea, we also find that higher subjective probabilities of either Social Security reform or of a recession within the next 10 years are associated with a lower probability of eligibility. However, those variables appear not to predict uncertainty in the expected Social Security benefit amount. But this could be due to the much smaller sample size in the analysis of the uncertainty about expected benefit amounts.

We describe associations between expectations and observable characteristics, but are unable to conclude whether the relationships are causal. Having panel data on individuals’ uncertainty about Social Security benefits—ideally on a larger sample—would enable us to make some progress in determining what factors are relevant when people form their expectations about future Social Security benefits. More specifically, it would be very valuable to have panel data on expectations, combined with data on the information that respondents receive between waves. Because collecting data on the changes in the information set between waves may be challenging, researchers could alternatively randomly provide “information treatments” to some respondents and not to others and evaluate how those treatments change expectations. In this context, it would also be of interest to have additional information on other potential determinants of uncertainty about benefits, such as the subjective distribution of a spouse’s claiming age and benefits.

In terms of the question of whether uncertainty about Social Security expectations influences portfolio choice, we faced the challenge that the uncertainty associated with the distribution of Social Security benefits unconditional on a claiming age may be endogenous, given that expectations about stock market returns may influence the subjective distribution of claiming ages. We mitigated this endogeneity problem by using the uncertainty associated with the distribution of Social Security benefits conditional on a claiming age. Our results suggest that individuals with more uncertainty about their future Social Security benefits tend to hold a smaller portion of their wealth in stocks. The magnitude of the estimated effect is likely an underestimate, because the uncertainty about Social Security benefits is bound to be smaller when measured conditional on a claiming age as opposed to when measured without such conditioning.

Admittedly, we cannot say for certain that the standard deviation of the conditional distribution in Social Security benefits is not correlated with unobservable variables that may influence portfolio choice. Specifically, it may be that expectations about Social Security benefits and stock market returns are correlated and that this correlation may influence portfolio choice. To find out, one would have to elicit the joint subjective distributions from survey respondents. To our knowledge, this has never been done. Future research will have to show whether eliciting the joint distribution is feasible, and whether, in our context, such correlation is empirically important.

Acknowledgments

This research was supported by a grant from the National Institute on Aging (P01AG008291 and R01AG20717). Delavande also acknowledges support form a NOVA Forum research grant. The authors thank the Health and Retirement Study (HRS) for providing the opportunity to include our module in the HRS Internet Survey. They also thank the participants of the workshop on “Subjective Beliefs in Econometric Models” held in Québec City in 2009, in particular Charles Bellemare and Chuck Manski, two anonymous referees and Luis Vasconcelos for their comments.

Appendix A: Proof of proposition 1

The proof is similar to that of Proposition 2 of Bodie and Samuelson (1989) . Consider first the no-uncertainty case. Denote L i ( x, B ) the leisure that satisfies the first-order condition with respect to L i given x and B . Let C i ( x, B ) denote the consumption that satisfies the first-order condition with respect to C i given x and B . Denote x ( B ) the optimal x given B . In the log-utility case, we get:

By the first-order-condition with respect to x , we know that V ( x ( B ), B ) = 0.

When we introduce uncertainty in Social Security benefits, the new first-order-condition with respect to x is: E b [ V ( x ( b ), b )] =0.

We can show that V ( x ( B ), b ) is concave in b for the class of utility functions considered in the proposition. Using Jensen inequality, we have:

Since E b [ V ( x ( b ), b )] is decreasing in x , we find that x * satisfying E b [ V ( x *, b )] = 0 is such that x * < x ( B ).

Appendix B: Fitting the subjective distributions of future Social Security benefits to parametric distributions

We follow closely Engelberg et al. (2009) to fit respondents’ answers to a parametric distribution. We use the subjective distribution provided in the unfolding bin screen for all the respondents who answered it. We make the following assumptions depending on the respondent’s answers:

• If the respondent uses one bin only, we assume that the subjective distribution has the shape of an isosceles triangle whose support is the bin.
• If the two bins are adjacent and each contains 10 balls, we assume that the subjective distribution has the shape of an isosceles triangle whose support is the union of the two bins.
• If the two bins are adjacent but contain unequal probability mass, we assume that the subjective distribution has the shape of an isosceles triangle. The support contains the bin with the higher probability mass entirely, and part of the bin with fewer probability mass to ensure that the probability allocated by the respondent within each bin is maintained.
• If the two bins are not adjacent (which is the case for 8 observations), we assume that the distribution is uniformly distributed within the bins with positive probability mass.
• If the respondent uses 3 bins or more, we assume that the subjective distribution is a generalized Beta distribution. Let T 0 = 0, T 1 ,…, T 7 be the thresholds for a given respondent, and let P j = P ( SS ≤ T j ), j = 0,…,6 denote the associated elicited cumulative distribution function of future Social Security benefits. The support of the beta distribution is assumed to be from zero to U = T 7 + T 1 , i.e. the right outer bin is assumed to be closed rather than open-ended and of similar width as the other bins. We find the parameters a and b of the beta distribution B (.; a, b , 0, U ) that best fit the answers of respondent i using nonlinear least squares. I.e, we solve: min μ , σ ∑ j = 0 6 ( P j − B ( T j ; a , b , 0 , U ) ) 2 .

Appendix C: Expected claiming age and distribution of claiming ages

We investigate which features of the distribution respondents report when asked for their point estimate of the age at which they expect to start receiving Social Security benefits. We focus the analysis on respondents with a point estimate between 60 and 70, and with at most 2 modes (97.1% of the data). We make two assumptions for this analysis:

• A1 The claiming age is a discrete random variable distributed between 60 and 70.
• A2 If respondents report the average, it is provided in terms of age in years (e.g., if the average of the distribution is 65.6, the respondent would report 65).

If a respondent allocates zero probability mass to the extreme bins (which are open-ended), we can compute exactly the mean, median and mode provided that A1 and A2 hold. For respondents who allocate a positive probability to the extreme bins, we can evaluate bounds for the mean, median or mode. For example, we compute a lower (upper) bound for the mean assuming that all the probability mass allocated to the interval “61 or less” is allocated to age 60 (61) and that all the probability mass allocated to the interval “68 or more” is allocated to age 68 (70). In some cases where the respondent allocated a small probability to the extreme bins, the upper and lower bounds for the mean are equal due to the rounding assumption posited in A2. Similarly, if a respondent allocates, for example, a 60% chance that the claiming age is “61 or less,” then the median and the mode can be 60 or 61. Overall, for 81% of the observations, we can compute the exact mode(s), median and mean. For the remaining 19%, we have a bound on at least one of these features of the distribution.

1 Social Security statements sent out to people every year also highlight the potential for Social Security reform, noting that “[Your] estimated benefits are based on current law. Congress has made changes to the law in the past and can do so at any time. The law governing benefit amounts may change because, by 2040, the payroll taxes collected will be enough to pay only about 74 percent of scheduled benefits.”

2 See Gollier (2002) , for example. In a model that distinguishes between risky and safe assets—holding all else the same—a person who faces uninsurable background risk should invest a smaller share in the risky asset. In the case we investigate here, the background risk comes from uncertainty about future Social Security benefits.

3 The elicitation method of the subjective distribution of replacement rates follows the method of Dominitz and Manski (2006) of asking respondents to first provide a minimum and maximum value, followed by questions about the chances that the replacement rate might be higher than a certain threshold. However, Guiso et al. (2009) only have one additional point on respondents subjective distribution while Dominitz and Manski elicited six additional points.

4 Monthly Social Security benefits depend on the claiming age for two reasons. First, given a fixed earnings history, claiming later is associated with a higher monthly benefit to compensate for the fact that the worker would receive the benefits for a shorter period. This adjustment of monthly benefits is designed to be actuarially neutral so that on average Social Security wealth is unaffected by the timing of claiming. Second, if an individual who claims later also retires later, then delaying claiming changes the worker’s earnings history and increases Social Security wealth for having contributed for a longer period. In our model, B × A individual’s Social Security wealth, which is increasing with the length of the individual’s working life.

5 The upper bounds on C 0 and L 0 ensure an interior solution for C i and L i , i=1,2 .

6 Spouses of age-eligible respondents are included irrespective of age.

7 The interview was postponed for one group of the sample to a later second phase because that group had previously been assigned (at random) to participate in another supplemental study. It would have had three HRS-related interviews within a few months of each other had the internet interview not been postponed.

8 The purpose of using two different elicitation formats was to evaluate the impact of the elicitation design on respondents’ answers (see Delavande and Rohwedder, 2008 ). Delavande and Rohwedder (2008) find that in the visual format the observed distributions tend to be more concentrated around the point estimate than the percent chance format. Additional experiments lead us to conclude that this is not due to anchoring toward the middle in the bins-and-balls format. In addition, 20 percent of the observations randomized into the percent chance format are lost due to violation of the monotonicity property of a cumulative distribution, and those 20 percent tend to be less educated, less healthy and less wealthy. We have therefore decided to focus our analysis on respondents randomized into the visual formats.

9 In our regressions, we exclude two respondents for whom education is missing. We also exclude from the analysis of uncertainty respondents who report a monthly Social Security benefit above $10,000 (11 respondents).

10 The percentages for the HRS 2006 are based on weighted data.

11 Both phase I and phase II respondents were asked to report the point estimate of their expected claiming age. The distribution of claiming ages was only administered to phase II respondents (N=1,021, with item non-response equal to 0.4%). All Phase II respondents were asked first their expected claiming age, and then their subjective distribution of claiming ages.

12 As mentioned earlier, the other half of the sample was also asked about their uncertainty about the Social Security benefit amount, but using a different survey design (a sequence of percent chance questions).

13 If all probability mass was allocated to a range between $300 and $450 in the first screen, respondents were provided a follow-up screen with 2 bins. If that initial range was between $450 and $600, the follow-up screen contained three bins. If the initial range was between $600 and $750, the new screen contained four bins. If the initial range was above $750, the new screen contained five bins.

14 Those in the first group were asked about their expected benefits first, followed by questions about when they would expect to claim these benefits. The second group, that is, those who were asked to condition on their expected claiming age when reporting their benefits expectations, naturally were first asked their expectations about claiming age, and then their expectations about Social Security benefits.

15 The HRS core survey asks about the point estimate of expected Social Security benefits conditional on claiming at the stated (point estimate of) expected claiming age.

16 The normal retirement age depends on the individual’s year of birth, and ranges from 65 to 67 in our sample.

17 A widowed person can claim as early as age 60.

18 For example, workers whose normal retirement age is 66 receive a benefit equal to 75% of the PIA if they claim on their 62nd birthday (e.g., Coile et al., 2002).

19 This may raise the concern that the middle bin attracts excessive probability mass due to anchoring bias. This is not the case as we found from experiments that we conducted in a different Internet survey ( Delavande and Rohwedder, 2008 ). We randomly administered to some respondents the same visual format as is done in the HRS Internet survey, and to other respondents a similar design where the third, rather than the middle bin is centered around the point estimate. We find that for both groups, most of the probability mass is attracted to the bin containing the point estimate, even though for one group this bin is not located in the middle.

20 Note that we find less uncertainty than Dominitz and Manski (2006) . They report that the cross-sectional median of the subjective interquartile range of yearly benefits is $7,140, while it is $1,296 in our data for the distribution unconditional on claiming age. This difference may be due to the fact that their sample is younger on average than ours (41.8 years old, in contrast with 55.9 years old). As we explain below, younger respondents report more uncertainty about future benefits.

21 We include categorical variables for both age and the distance to the expected claiming age. While younger respondents have on average a larger distance to the expected claiming age, there is heterogeneity in this variable within age categories, in particular among the 50-to-59-year-olds.

22 To compute the standard deviation of the distribution of claiming ages, we treat claiming ages as a discrete random variable distributed between 60 and 70. Those who report a positive probability to the interval “61or less” are assumed to have allocated half of this probability to age 60, and half of it to age 61. We make similar assumptions for those who report a positive probability for the interval “68 or more.”

23 In 2006, respondents younger than 62 were asked their expected benefits if they were to claim at age 62 and if they were to claim at their normal retirement age. Of those who were asked both questions, 53% correctly provided a lower benefit amount for claiming at age 62 than for claiming at the normal retirement age, 10% made a mistake (that is, they reported an equal or higher amount at age 62 than at the normal retirement age), and 37% answered “don’t know” to at least one of the questions about the expected benefit amounts.

24 The standard deviation of the distribution of the benefit amount unconditional on eligibility is given by the product of the probability of being eligible and the standard deviation of the distribution of the benefit amount conditional on eligibility.

25 All components of wealth in the HRS are measured at the household level. Total wealth is defined as the sum of the following assets: housing, other real estate, transportation, IRAs, stocks and stock mutual funds, checking and savings, CDs, bonds, and other assets, minus all debt (mortgages, home equity loans, and any other debt). With respect to IRAs, respondents are asked what fraction of the balance is invested in stocks. We applied these reports to the respective IRA balances to recover the amount held in stocks.

26 Table 3 excludes respondents who report a zero probability of receiving Social Security in the future. Those have a zero mean and a zero standard deviation. Including them, while adding an indicator for reporting a zero probability, does not change our results.

27 In an alternative specification, we have also included the measures of Social Security knowledge described earlier to proxy for financial literacy. They were imprecisely estimated and other coefficients were unaffected by their inclusion.

Contributor Information

Adeline Delavande, RAND Corporation and Universidade Nova de Lisboa, Economist and Assistant Professor.

Susann Rohwedder, RAND Corporation and NETSPAR, Senior Economist, 1776 Main Street, CA 90407, Santa Monica, Tel: 310-393.0411, Fax: 310-451.7084, gro.dnar@rnnasus .

• Bernheim D, Levin L. Social Security and Personal Saving: An Analysis of Expectations. American Economic Review; Papers and Proceedings of the Hundred and First Annual Meeting of the American Economic Association; 1989. pp. 97–102. [ Google Scholar ]
• Bodie Z, Samuelson W. Labor Supply Flexibility and Portfolio Choice. NBER Working Paper No 3043 1989 [ Google Scholar ]
• Bodie Z, Merton R, Samuelson W. Labor Supply Flexibility and Portfolio Choice in a Life-Cycle Model. Journal of Economic Dynamics and Control. 1992; 16 (3-4):427–449. [ Google Scholar ]
• Bottazzi R, Jappelli T, Padula M. The Portfolio Effect of Pension Reforms: Evidence from Italy. Journal of Pension Economics and Finance. forthcoming. [ Google Scholar ]
• Bottazzi R, Jappelli T, Padula M. Retirement expectations, pension reforms, and their impact on private wealth accumulation. Journal of Public Economics. 2006; 90 :2187–2212. [ Google Scholar ]
• Campbell J. Household Finance. The Journal of Finance. 2006; 64 :1553, 1604. [ Google Scholar ]
• Delavande A, Rohwedder S. Eliciting Subjective Expectations in Internet Surveys. Public Opinion Quarterly. 2008; 72 :866–891. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Dominitz J, Manski C. Measuring Pension-Benefit Expectations Probabilistically. Labour. 2006; 20 :201–236. [ Google Scholar ]
• Dominitz J, Manski C. Using Expectations Data to Study Subjective Income Expectations. Journal of the American Statistical Association. 1997; 92 :855–867. [ Google Scholar ]
• Engelberg J, Manski C, Williams J. Comparing the Point Predictions and Subjective Probability Distributions of Professional Forecasters. Journal of Business and Economic Statistics. 2009; 165 :146–158. [ Google Scholar ]
• Gollier C. What Does Theory Have to Say About Household Portfolios? In: Guiso G, Haliassos M, Jappelli T, editors. Household Portfolios. MIT Press; 2002. pp. 27–54. [ Google Scholar ]
• Guiso L, Jappelli T, Padula M. Pension Risk, Retirement Saving and Insurance. CSEF Working Paper No 223 2009 [ Google Scholar ]
• Gustman AL, Steinmeier TL. What People Don’t Know About Their Pensions and Social Security: An Analysis Using Linked Data From the Health and Retirement Study. In: Gale WG, Shoven JB, Warshawsky MJ, editors. Public Policies and Private Pensions. Washington, D.C.: Brookings Institution; 2004. [ Google Scholar ]
• Lusardi A, Mitchell O. Financial Literacy and Retirement Preparedness. Evidence and Implications for Financial Education Programs. Business Economics. 2007:35–44. [ Google Scholar ]
• Merton RC. Continuous-Time Finance. Blackwell Publishers; Cambridge, MA: 1990. [ Google Scholar ]
• U.S. Social Security Administration. Annual Statistical Supplement. 2009 www.socialsecurity.gov/policy/docs/statcomps/supplement/2009/

Advertisement

Social cybersecurity: an emerging science

• S.I. : Social Cyber-Security
• Open access
• Published: 16 November 2020
• Volume 26 , pages 365–381, ( 2020 )

Cite this article

You have full access to this open access article

• Kathleen M. Carley   ORCID: orcid.org/0000-0002-6356-0238 1  

35k Accesses

67 Citations

15 Altmetric

Explore all metrics

With the rise of online platforms where individuals could gather and spread information came the rise of online cybercrimes aimed at taking advantage of not just single individuals but collectives. In response, researchers and practitioners began trying to understand this digital playground and the way in which individuals who were socially and digitally embedded could be manipulated. What is emerging is a new scientific and engineering discipline—social cybersecurity. This paper defines this emerging area, provides case examples of the research issues and types of tools needed, and lays out a program of research in this area.

Similar content being viewed by others

Social Cyber-Security

Defining Cybercrime

Avoid common mistakes on your manuscript.

In today’s high tech world, beliefs opinions and attitudes are shaped as people engage with others in social media, and through the internet. Stories from creditable news sources and finding from science are challenged by actors who are actively engaged in influence operations on the internet. Lone wolfs, and large propaganda machines both disrupt civil discourse, sew discord and spread disinformation. Bots, cyborgs, trolls, sock-puppets, deep fakes, and memes are just a few of the technologies used in social engineering aimed at undermining civil society and supporting adversarial or business agendas. How can social discourse without undue influence persist in such an environment? What are the types of tools and theories needed to support such open discourse?

Today scientists from a large number of disciplines are working collaboratively to develop these new tools and theories. There work has led to the emergence of a new area of science—social cybersecurity. Herein, this emerging scientific area is described. Illustrative case studies are used to showcase the types of tools and theories needed. New theories and methods are also described.

1 Social cybersecurity

In response to these cyber-mediated threats to democracy, a new scientific discipline has emerged—social cybersecurity. As noted by the National Academies of Science NAS ( 2019 ): Social cybersecurity is an applied computational social science with two objectives

“characterize, understand, and forecast cyber-mediated changes in human behavior and in social, cultural, and political outcomes; and build a social cyber infrastructure that will allow the essential character of a society to persist in a cyber-mediated information environment that is characterized by changing conditions, actual or imminent social cyberthreats, and cyber-mediated threats.”

Social cybersecurity is both a new scientific and a new engineering field. It is a computational social science with a large foot in the area of applied research. Drawing on a huge range of disciplines the new technologies and findings in social cybersecurity have near immediate application on the internet. The findings and methods are relevant to policy makers, scholars, and corporations.

Social cybersecurity uses computational social science techniques to identify, counter, and measure (or assess) the impact of communication objectives. The methods and findings in this area are critical, and advance industry-accepted practices for communication, journalism and marketing research. The field itself has a theory, application, and policy component. The methods build on work in high dimensional network analysis, data science, machine learning, natural language processing and agent-based simulation. These methods are used to provide evidence about who is manipulating social media and the internet for/against you or your organization, what methods are being used, and how these social manipulation methods can be countered. They also support cyber diplomacy (Goolsby 2020 ).

Social cybersecurity uses computational social science techniques to identify, counter, and measure (or assess), the impact of influence campaigns, and to identify and inoculate those at risk against such campaigns. The methods and findings in this area are critical, and advance practices for intelligence and forensics research. These methods also provide scalable techniques for assessing and predicting the impact of influence operations carried out through social media, and for securing social activity on the internet and mitigating the effects of malicious and undue influence. As such they are critical for creating a more secure and resilient society.

Influence campaigns vary widely, and who is at risk in part depends on those conducting the influence campaign and in part on the context. For example, in our research we found that influence campaigns appearing to come from state-level actors during the elections in Western Europe and the US from 2016 to 2020 were often aimed at minorities. For example, they targeted women, ethnic minorities, and the LGBQT community. In contrast, in India as COVID-19 ramped up internal non-state groups launched anti-Muslim campaigns. As movies like the Black Panther and Captain Marvel were released individual’s launched campaigns against the movies. In the elections in the Asia Pacific region influence campaigns often take the form of promoting pro-China candidates. Many influence campaigns are aimed at specific individuals trying to recruit them to a new cause, or engage them in insider threat activity.

Social cybersecurity is distinct from cybersecurity. Cybersecurity is focused on machines, and how computers and databases can be compromised. In contrast, social cybersecurity is focused on humans and how these humans can be compromised, converted, and relegated to the unimportant. Where cybersecurity experts are expected to understand the technology, computer science, and engineering; social cybersecurity experts are expected to understand social communication and community building, statistics, social networks, and machine learning. Social cybersecurity is also distinct from cognitive security. Cognitive security is focused on human cognition and how messages can be crafted to take advantage of normal cognitive limitations. In contrast, social cybersecurity is focused on humans situated in society and how the digital environment can be manipulated to alter both the community and the narrative. Where cognitive security experts are expected to understand psychology, social cybersecurity experts are expected to have a broader social science expertise.

In our research we have found that there is some work in social cybersecurity that draws on most scientific fields. In a recent study of the field, we identified 1437 papers up through 2019. Each journal was coded by the dominant scientific fields that it is associated with. The results was a set of 43 disciplines. In Fig.  1 we show the discipline to discipline network where the links indicate the number of articles that draw on both disciplines. The size of the nodes indicate the number of articles associated with that discipline.

Network diagram of the interdisciplinary nature of the field of social cybersecurity. Nodes are disciplines and are sized by number of articles. Links are number of articles associated with both disciplines

As seen in Fig. 1 , social cybersecurity is very much a computational social science, with a strong interdisciplinary focus drawing on the social and communication and computer sciences. The dominant methods are social network analysis, data mining, and artificial intelligence (which includes language technologies and machine learning). The areas that have dominated research in this area are largely ones that draw on theories from diverse disciplines.

Within social cybersecurity artificial intelligence is coupled with social network analysis to provide new tools and metrics to support the decision maker. Recent research in social cybersecurity is enabling new tools to support research methodology and metrics-based decision making for communicators. The following case studies highlight the types of research findings made possible in the area of social cybersecurity using these new tools. After presenting these case studies we then turn to a discussion of a new orienting approach for doing research in social cybersecurity, referred to as the BEND framework. Collectively, these items provide a glimpse into the core of this emerging scientific field.

What are the dominant themes in social cybersecurity? As can be seen in Fig.  2 , the dominant research area currently is disinformation. This is followed by research on user behavior and networks on the web, and then research on politics and democracy. In Fig. 2 each node represents a research topic and the size of the node reflects the number of articles on that topic. There are two caveats, first, the size of the disinformation node is growing rapidly with all the new papers related to COVID-19 and the elections. Second, the reader may wonder why privacy does not appear. This is because privacy was viewed as a separate field unto itself and papers in that area were not include in the analysis.

Research topic areas in social cybersecurity

2 Case study 1: building community in social media

In Ukraine there were a group of young men sending out provocative images of women. They didn’t know each other they were just posting images they liked. Bots were used in an influence campaign to send out tweets mentioning each other and multiple of these young men at once. This led the men to learn of others who, like them, were sending out these images. They formed an online group—a topic oriented community. Once formed, the bots now-and-then tweeted information about where to get guns, ammunition, and how to get involved in the fight. Why did this work?

The cyber landscape is populated by topic oriented communities—groups of actors all communicating with each other about a topic of interest. Each actor can be in many topic oriented communities. Actors can be people, bots, cyborgs (person with bot assistance), trolls (person seeking to disrupt, a corporation or government account using a fake persona and often engaging in hate speech and identity bashing), and so forth. Members of a topic oriented community are loosely connected by the fact that they interact with each other. For example, they might friend, follow, retweet, mention, reply, quote or like each other. Some actors will be opinion leaders, some will have a disproportionate ability to get messages to the community (super-spreaders), some will be highly involved in the mutual give and take of an on-going discussion (super-spreaders), some will just be lurking on the sidelines. The members of topic oriented communities are also loosely connected because they are sending or sending or receiving messages about the same topics. For example, they are all discussing the Army-Navy game. Some actors will be more actively engaged and send more messages. Topic oriented communities range in size and how they are organized; e.g. plane spotters is a vast community that is only slightly connected. Through new tools and research methodologies that measure communication impacts via social media, it is now possible to measure and visualize data to demonstrate topic oriented communities that become overly connected become echo-chambers. (Note: An echo-chamber is a pathologic form of a topic oriented community in which the level of connection is extremely dense and the topic highly shared and narrow.) Messages sent within echo-chambers reach all quickly, and such groups can often be readily excited to respond emotionally rather than rationally to outside information.

For Ukraine bots were used to send communications which basically introduced these young men to each other through the use of @mentions. These bots also sent provocative images. The young men then began to follow each other forming a topic oriented community. In Ukraine influencers created/controlled the bots that conducted a “build” campaign to misinform (engendering social connections between the young men by mentioning them together). At the same time, they conducted an “enhance” campaign by rebroadcasting some images and pointing to others and an “excite” campaign with new positive language. Once the group was established, a “distort” campaign appeared bringing in information relative to the revolution.

For additional details on this case study see (Benigni et al. 2019 ).

3 Case study 2: increasing communicative reach in social media

Syrian expats and sympathizers with ISIS were engaged in social media conversations. This included listening to the preaching’s of a prominent Imam. A group of actors infiltrated this group and redirected attention to a site collecting money for the children of Syria. How was this done?

In social media, your followers may not receive your messages, or your messages may not be prioritized so that they appear prominently to those concerned with your messages. Social media platforms use your social network position (how you are connected to others), and the content of your message, to decide who to recommend your message to, when, and in what order. Who you mention in posts, which hashtags you use, whether you use memes or link to YouTube videos, the frequency with which you post, the number of others who follow you or like your posts, all impact whether your message is prioritized.

In this Syrian ex-pat community influencers created/controlled a social influence bot, the Firibi gnome bot. This set of bots was used to conduct a sophisticated influence campaign. Multiple copies of this bot were released which proceeded to send out messages mentioning each other and so engaged in a “build” campaign to misinform. The results was a topic oriented community of bots, which meant that messages from any one bot would be recommended to others interested in similar topics. Then these bots started following retweeting messages from an Imam, who may not have been aware of this activity. This “boosted” the social influence of the Imam, and “engaged” the bot with the community. Since the Imam was a super-spreader, this also meant that messages from the Firibi gnome would be prioritized to the Imams followers. Then, the Firibi gnome bot engaged in an “enhance” campaign and started sending messages recommending the charity website. This message was then prioritized.

For additional details on this case study see (Benigni et al. 2017 ).

4 Case study 3: conspiracies in social media

As COVID-19 spread so to dis disinformation regarding the pandemic. Thousands of disinformation stories were spread focusing on false cures and prevention techniques, false characterization of government response, and claims of leaders having COVID even when they did not. Throughout a number of conspiracy stories appeared and began to gain traction. How was this done?

In social media, messages gain traction through amplification and saturation. A message from a single actor or site can be amplified and spread by bots and trolls. This means those messages will get shared disproportionately and may even be made to trend. The more shared, the more the story gets prioritized to individuals who do not pay attention to the original source but may be following users who follow those who follow the source. Further, in social media the presence of a story on two or more platforms does not mean that the story has been independently validated. Indeed stories often appear on one platform, and bots and trolls are used to push it to other platforms. Marketing firms hired to spread disinformation can place the same story simultaneously in different forms on each of the platforms. Twitter, e.g., is often used to garner attention to YouTube videos and to promulgate stories that first appear in Blogs or on Facebook. The more platforms a story appears on, the more it saturates the digital space, the more real it seems, particularly when it is accompanied by images and videos and supported by celebrities and authorities.

A number of conspiracies surrounded COVID-19. One is that the virus was created in a US lab and carried to Wuhan by US soldiers engaged in a war game, and another is that it was created by Bill Gates and then spread as step 1 in a plan to create a new world order. Stories regarding some of these conspiracies appeared on Chinese state sponsored media. Bots surrounding these media then retweeted the stories—thus amplifying the reach. Related stories and even a “plandemic” video were released on multiple media e.g., Facebook, YouTube, Twitter and Instagram. Bots further amplified these messages or sent messages with the related URLS. Trolls, employed hate speech to denigrate those that tried to counter the conspiracy messages, were implicated by the conspiracy messages, or that were anticipated to not believe the conspiracy messages. The same conspiracy stories, often providing additional details, also appeared on the purported “fake news” news-sites which are websites that purport to be news agencies but either are not news sites or have dubious editorial procedures and are known for spreading disinformation. Large numbers of bots surround these sites and would send out messages with URLS to these sites. Even larger numbers of bots retweeted messages referencing these sites. The result was a topic oriented community of conspiracy theorists, bots, and trolls around various conspiracy thrusts, an increase in the number of conspiracies that were spreading, the re-appearance of conspiracy stories even if they were banned, and increasingly elaborate conspiracies such as the new world order.

For additional details on this case study see (Huang, 2019 ).

5 The BEND framework

The foregoing case studies indicate the types of issues that need to be considered by a social cybersecurity researcher or practitioner. They also suggest the need for new technologies such as those to identify disinformation, bots, trolls, cyborgs and memes. Finally they point to the need for new theories to make sense of the way in which influence plays out in social media. One such transdisciplinary theory is referred to as the BEND framework.

Influence campaigns are often described in terms of the 4Ds—distract, distort, dismay and disrupt (Nimmo, 2015 ). These are often used to describe information operations by Russia. However, as the three case studies illustrate influence operations don’t involve just messages with distract, distort, dismay or disrupt campaigns. Our research suggests that a broader understanding of information maneuvers is needed. Specifically, information campaigns that are successful typically impact both community and narrative. That is maneuvers are conducted that alter both who is communicating with whom as well as what is being communicated. Further, the 4Ds are essentially negative maneuvers; that is, they are problem creation not problem solution maneuvers. In the Ukraine—images were used to excite, in Syria—messages were used to explain, in COVID-19 explain and enhance messages were used. While in each case these are not the only kind of messages used, the point is, there was more than just the 4Ds. The.

BEND framework argues that influence campaigns are comprised of sets of narrative and structural maneuvers, carried out by one or more actors by engaging others in the cyber environment with the intent of altering topic-oriented communities and the position of actors within these communities. A topic oriented community is a group of actors who are more or less talking to each other about more or less the same thing. This engagement with the topic oriented community is often assisted by or carried out by bots, trolls and cyborgs in addition to human users. This engagement is aimed at manipulating either or both the narrative (what is being talked about) and the community (who is talking to whom). Bots, trolls, cyborgs and humans engage with others in cyberspace in ways designed to, and send messages that are constructed to, take advantage of three things: the technology, the mind and emotions, and the world view. Technology: these activities are designed to exploit the algorithms that prioritize the order in which messages are presented and the recommendation algorithms so that messages selected by the perpetrator appear first, often, and trend, and the goods, services, urls, and actors they mention are recommended to the readers. Mind & Emotion: these activities are designed to exploit natural human biases and reflexes such as confirmation bias, escalation of commitment, and the fear or flight reflex. World View: these activities are designed to make use human social cognition, the set of heuristics we use to make sense of vast quantities of data in terms of group—such as the generalized other and stereotyping.

In social cybersecurity, theories and methods go hand-in-hand. Hence associated with the BEND theory is a methodology for empirically assessing which maneuvers are being used and for measuring the impact of social media communication research, planning, and objectives (Beskow and Carley 2019 ). The BEND framework thus has associated with it a set of methods and tools for looking at who engaged in what information maneuvers directed at whom with what impact. The BEND framework characterizes communication objectives and so the maneuvers into 16 objectives such that 8 are aimed at shaping the social networks of who is communicating with whom and 8 are aimed at shaping the narrative. For the social network, there are four positive objectives (the four B’s) and four negative objectives (the four N’s). Similarly, for shaping the narrative there are four positive objectives (the four E’s) and the four traditional negative objectives (the four D’s). These are described in Table  1 .

The BEND framework is the product of years of research on disinformation and other forms of communication based influence campaigns, and communication objectives of Russia and other adversarial communities, including terror groups such as ISIS that began in late December of 2013. It draws on both findings regarding the communication objectives and tactics of adversarial actors adversarial (Benigni et al. 2017 ; Lucas and Nimmo 2015 ; Manheim 1994 ), political influence (Howard and Kollanyi 2016 ; Howard et al. 2018 ; Huckfeldt and Sprague 1995 ), marketing (Webster 2020 ), psychology (Sanborn and Harris 2013 ). The BEND Framework addresses these communication objectives and tactics from a transdisciplinary perspective. Early evidence suggests that excite, enhance, dismay, and distort may be the most common communication objectives used to spread disinformation.

The BEND framework is more than a description of the maneuvers shown in Table 1 . It begins with identifying the types of user who is conducting a maneuver or is targeted by such a maneuver. Actors are characterized by whether they are bots, trolls, news agencies, government actors, celebrities or are influential super-friends (those with a high number of reciprocated ties in social media), super-spreaders (those with a high number of others who they reach with their messages e.g. a large number of followers etc.) or are otherwise influential in social media (e.g. send a large number of messages). Actors being targeted can be topic oriented communities or individual actors. Then each of the messages are characterized for which of the 16 maneuverers the message is consistent with using a set of discreet measures. Finally, impact in terms of change in the target is assessed from a social network perspective and content perspective exploring how over time the target has changed.

Associated with the BEND maeuvers are a series of measures and indicators for each of the objectives. These have been operationalized, are now part of the ORA-PRO social media tools, and have been tested on Twitter data. They were used in assessing data during various NATO exercises, Naval exercises, elections, and disasters. We find that in many cases complex influence campaigns involve using multiple BEND objectives as was described in the three case studies.

6 Using social network analysis and artificial intelligence

One of the key tools in social cybersecurity is high dimensional dynamic social network analysis. Social network analysis is the analysis of who interacts with whom. Network techniques have long been used in intelligence for identifying groups and tracking adversarial actors and by marketers for identifying key informants and opinion leaders. With social media such techniques have been expanded to enable scalable solutions for massive data that take into account multiple types of relations among actors as well as relations among resources, ideas and so forth. Today such high dimensional dynamic network techniques underlie social media analysis. The two interaction networks—such as (1) who likes or retweets whom and (2) the content of the messages—are treated as networks. The techniques to identify these interactions are embedded in ORA-PRO and are used for identifying topic-groups and the influential actors within these groups; the depth of this data is not possible with other off-the-shelf analysis tools. Running social network techniques on social media provides indicators that can then be used in machine learning tools to identify actors and messages of interest such as bots, cyborgs, and trolls.

Artificial intelligence (AI) techniques, particularly machine learning and natural language processing techniques are also key tools in social cybersecurity. AI, and particularly machine learning (ML), are often pointed to as force multipliers in dealing with the vast quantity of digital data available today. Such technologies are clearly of value; however, they are not the panacea envisioned. The problems faced by the military in social cyberwar are continually changing and often occur only once; thus, new technique for responding are continuously needed. Further, current AI and ML techniques are often focused on easily measured data rather than the more volatile socio-political-cultural context.

Language technologies are used for translation, sentiment, and stance detection. Most sentiment tools simply inform the reader if a message containing a word of interest is positive or negative, which often has no relation to the sentiment about the word of interest. We find that as much as 50% of the time the sentiment toward the word of interest is the opposite as the sentiment of the message as a whole. Consider the sentiment toward U.S. in the message—“I hate Russian interference in social media and treatment of the U.S. as evil”. The sentiment of the message as a whole is negative; but it is positive toward the U.S.. In contrast, the NetMapper system used with the BEND framework identifies the sentiment about the word of interest, and measures a set of subconscious CUES in the message to assess the sender’s emotional state.

Machine learning techniques are frequently used to identify bots, false statements, and message on particular topics. An example is BotHunter. Such tools can rapidly identify the likelihood that potential actors are Bots. This can indeed support analysis and help a communicator understand an adversary’s communication objectives. However these, tools that are based on “supervised” learning have a limited shelf life. First, they require large training sets. Training sets need to be created by humans tediously coding messages and the senders of messages into categories required for the AI tool. Today, bots are evolving faster than are the tools to find them in large part because it takes too long to create training sets. Training sets are often biased—e.g., sentiment training sets are biased toward lower middle-class ways of expressing sentiment in English. The AI tools themselves give probability scores and no explanation on why they reached the conclusion they did. Bot detection tools often disagree because the tools were “trained” differently—leaving the ultimate decision in the hands of the analyst. These factors reduce how long these technologies will be useful for and in what contexts. Today’s technology advances are being made in developing AI techniques that do not require massive training sets and that provide explanations—BotRecommender is such a tool.

For disinformation, the issues are legion and there are many types of disinformation as is illustrated in Table  2 . Fact-checking tools using humans or human-AI teams are providing valuable guidance but so far take a long time to determine if a story contains an inaccuracy. Assessing intent is difficult—was the sender intentionally trying to deceive (disinformation) or were they just mistaken (misinformation). Many disinformation campaigns are not based on inaccurate facts, but on innuendo, fights of illogic, reasoning from data taken out of context, and so on. Many times, stories labeled as disinformation are simple alternative interpretations of facts. AI only helps for some types of disinformation. It is less useful the more unique the storyline, and the faster the story spreads.

AI techniques are only useful as part of the toolkit. AI can support classifying messages by BEND objectives, or the perpetrators into types such as bots, trolls, and news-agents. The BEND framework and associated tools, some of which employ AI, can be used to assess how communications are spreading and measure the impact. For example, MemeHunter was used to identify an influence campaign from Russia engaged with a dismay objective that implied that compared to Russia, NATO was weak as the head of many countries defense were women, not a strong male military leader. This meme was spread by bots and humans alike.

7 Research directions

Social cybersecurity is an exciting and emerging field. The BEND framework and so the associated theory and methods, BotHunter, MemeHunter and other such tools, are examples of the kind of work central to this area. However, much remains to be done. Indeed there are seven core research areas.

Social Cyber-Forensics: Social cyber-forensics is concerned with identifying who is conducting social cybersecurity attacks. Often the concern is with the type of actor rather than the specific actor. Further, this can involve cross platform assessment with the need to track down the source of information. New ways to track and build linkages at scale are needed.

Information Maneuvers: The key here is to understand the strategies used to conduct an attack and the intent of those strategies. Can we, for example, expand upon BEND to identify sets of maneuvers that are consistently used together or in a particular order to effect a particular impact? Improved abilities to detect maneuvers, and provide early warning that an attack is starting, are needed—particularly cross platform.

Motive Identification: The goal here is to understand what the perpetrators motive is. Why is the attack being conducted? Multiple motives have been seen. These include conducting influence campaigns: for fun, to create havoc, to polarize society, to alleviate boredom, for money, to polarize society, to market goods or services, to gain personal influence, and to generate community. There are likely to be other reasons as well. Being able to identify and track motive at scale and quickly is an important area for new research.

Diffusion: In this area the objective is to trace, and even predict, the spread of an influence campaign. A sub-aspect of this is to trace, and even predict, the movement of the components of a campaign such as people, ideas, beliefs, memes, videos, and images. Tracing the attackers and the impact of the attack across and through multiple social media is key. Live monitors that suggest when diffusion is about to explode, peak, and peter out. Improve theories of and methods for monitoring diffusion, particularly cross-platform are needed.

Effectiveness of Information Campaigns: The goal of this area is to quantify the effectiveness of the social cyber-security attack. This includes both the short term and the long term impact. It also involves creating improved measures of impact – such as polarization or mass-hysteria – rather than the traditional measures of reach such as number of followers, likes, and recommendation. While some measures cannot be done in real time, real time estimates of potential impact would be of value. New theories about impact and effect, as well as new techniques to measure effect are needed.

Mitigation: There are two related goals here. The first is to understand how a social cybersecurity attack be countered or mitigated. The second is to understand how communities can become more resilient to attacks. Many different avenues of research can be pursued here. Some examples are use of agent based models to assess the relevant impact of interventions, scalable techniques for teaching critical thinking for social media, and basic research on the characteristics of resilient communities. New empirical results, transdisciplinary theories, models and ways of measuring resilience and mitigation in this space are needed.

Governance: The objective here is to understand what policies and laws are needed so the people can continue to use the internet without fear of undue influence, so that an informed democracy can survive. This is a key area as it brings together issue of legality, rights, and education. This area needs to bring together all the diverse perspectives and diverse knowledge to develop actionable governance.

8 Conclusion

As noted, social cybersecurity is an emerging scientific and engineering discipline. While there are thousands working in this space, more research and more coordination of that research is needed. Work in this area began as interdisciplinary and is becoming transdisciplinary. Two words of caution. First, for individuals new to the area it is easy to come to the conclusion that little is known and that only a few individuals are working in this area. There are several reasons for this. First the research is spread across hundreds of venues with no one conference or journal being dominant. Second there is some research in most disciplines, but in each of the extant disciplines this is a fringe area. What we have found is that most researchers in this area do now know of others outside their own group. Often faculty in this area don’t even know of others in their own university. Greater outreach and ways of collaborating and coordinating across groups is needed. This is beginning to happen. In Fig.  3 , the collaboration network based on who co-authors with whom, for 2018 and late 2019 is shown. As can be seen the central core has grown, and there are now links where none existed before. These growth is largely due to the Department of Defense Minerva program and the Knight foundation, both of which began to support research in social cybersecurity and to encourage collaboration.

Evolving co-authorship network. The top image shows the central core in 2018 and the bottom image the central core in 2019. Each node is an author and the links are weighted by the number of papers those two authors co-authored

Second, it is easy to think of this area as one where computer science and artificial intelligence will provide the solutions. Artificial intelligence solutions often aim at the easy things such as fact checking and currently require large levels of training data. But, this is a fast moving area where training sets are difficult to come by and are often out of data by the time they are created. It is easy to fall prey to stories that claim success because the mined extremely large data sets. But this fails to recognize that what is being discovered is the mean behavior and that most social change and social activity is on the fringe, and in the margins. What both approaches fail to recognize is that at its heart, social cybersecurity is about people as social beings, and it is people as social beings that are impacting and being impacted. To be sure artificial intelligence and data science are critical to this area; however, they should be in supporting positions not the drivers sear. What is needed is socially informed, social human being led computational social science.

Benigni M, Joseph K, Carley KM (2017) Online extremism and the communities that sustain it: detecting the ISIS supporting community on Twitter. PLoS One 12(12):e0181405

Article   Google Scholar  

Benigni M, Joseph K, Carley KM (2019) Bot-ivistm: assessing information manipulation in social media using network analytics. In: Agarwal N, Dokoohaki N (eds) Emerging research challenges and opportunities in social network analysis and mining. Springer, Cham

Google Scholar  

Beskow DA, Carley KM (2019) Social cybersecurity: an emerging national security requirement, military review, March–April 2019—see https://www.armyupress.army.mil/Journals/Military-Review/English-Edition-Archives/Mar-Apr-2019/117-Cybersecurity/b/

Goolsby R (2020) Developing a new approach to cyber diplimacy. Future Force 6(2):8–15

Howard PN, Kollanyi B (2016) Bots, # strongerin, and # brexit: computational propaganda during the uk-eu referendum. Available at SSRN 2798311

Howard PN, Woolley S, Calo R (2018) Algorithms, bots, and political communication in the US 2016 election: the challenge of automated political communication for election law and administration. J Inform Tech Polit 15(2):81–93

Huang B (2019) Learning User Latent Attributes on Social Media. Ph.D. Thesis, Institute for Software Research, Carnegie Mellon University

Huckfeldt RR, Sprague J (1995) Citizens, politics and social communication: information and influence in an election campaign. Cambridge University Press, Cambridge

Book   Google Scholar  

Lucas E, Nimmo B (2015) Information warfare: what is it and how to win it. CEPA Infowar Paper 1

Manheim JB (1994) Strategic public diplomacy and American foreign policy: the evolution of influence. Oxford University Press, New York

National Academies of Sciences, Engineering, and Medicine (2019) A decadal survey of the social and behavioral sciences: a research agenda for advancing intelligence analysis. The National Academies Press, Washington, DC. Ch. 6

Nimmo B (2015) Anatomy of an info-war: how Russia’s propaganda machine works, and how to counter it. Central European Policy Institute, 15

Sanborn FW, Harris RJ (2013) A cognitive psychology of mass communication. Routledge, New York

Webster T. 8 surprising Twitter statistics that will help you get more engagement. https://postcron.com/en/blog/8-surprising-twitter-statistics-get-more-engagement/ . Accessed 2/2020

Download references

Acknowledgements

This paper is the outgrowth of research in the center for Computational Analysis of Social an Organizational Systems (CASOS), and the center for Informed Democracy and Social-cybersecurity (IDeaS) at Carnegie Mellon University. This work was supported in part by both centers, the Knight Foundation, and the Office of Naval Research through the Minerva program N00014-16-1-2324 and the Office of Naval Research N000141812108 and N00014182106. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Knight Foundation, the Office of Naval Research or the U.S. government.

Author information

Authors and affiliations.

Center for Informed Democracy and Social Cybersecurity, Carnegie Mellon University, Pittsburgh, PA, USA

Kathleen M. Carley

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Kathleen M. Carley .

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Carley, K.M. Social cybersecurity: an emerging science. Comput Math Organ Theory 26 , 365–381 (2020). https://doi.org/10.1007/s10588-020-09322-9

Download citation

Published : 16 November 2020

Issue Date : December 2020

DOI : https://doi.org/10.1007/s10588-020-09322-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Social cybersecurity
• Social network analysis
• Dynamic network analysis
• Social media analytics
• Find a journal
• Publish with us
• Track your research

Social Security

Disability research.

Empowering people with disabilities through research, demonstrations, and employment support

What's New?

Research & Demonstrations

Temporary initiatives to identify services, supports, and policies to support people with disabilities.

• Interventional Cooperative Agreement Program (ICAP)
• Promoting Work through Early Intervention Project (PWEIP)
• Retaining Employment and Talent After Injury/Illness Network (RETAIN)
• Accelerated Benefits
• Benefit Offset - Four-State Pilot
• Benefit Offset - National Demonstration (BOND)
• Disability Program Navigator (DPN)
• Homeless Outreach Projects & Evaluation (HOPE)
• Mental Health Treatment Study (MHTS)
• Promoting Opportunity Demonstration (POD)
• Promoting Readiness of Minors in SSI (PROMISE)
• State Partnership Initiative (SPI)
• Supported Employment Demonstration (SED)
• Technical Expert Panels (TEPs)
• Youth Transition Demonstration (YTD)

Articles, reports, evaluations, briefing papers, and other studies related to disability, work, and related topics.

• Analyzing Relationships between Disability, Rehabilitation, and Work (ARDRAW)
• Beyond Benefits Study
• Quick Disability Determinations (QDD)
• Ticket to Work Program Evaluation
• Cardiovascular Report
• Consultative Examination (CE) Study and Consultative Exam Baseline Study
• Disability Determination Process Small Grant Program (DDP) Final Report
• Employment Network Payment Structure Evaluation Report
• Pediatric Medical Unit (PMU)
• Quick Disability Determinations (QDD) Bias Investigation Project
• SSI Youth Community Based Services and Supports
• SSI Youth Employment Evidence Report
• TANF-SSI Disability Transition Project
• Ticket to Work Evaluation
• Understanding Experiences Study
• Vocational Expert Study Panel
• WIPA Service Model Analysis Report

Information released by SSA for the general public to use in statistical analyses.

• Disability Analysis File (DAF) - Public Use File
• Disability Analysis File (DAF) - Restricted Access File
• National Beneficiary Survey (NBS)
• National Survey of SSI Children and Families (NSCF)
• Occupational Information System (OIS) Project
• Work Disability Functional Assessment Battery (WD-FAB) Research Study

Employment Support

Policies to help beneficiaries work by protecting cash and medical benefits.

Programs and services to help beneficiaries return to work and succeed in the labor force, including the Ticket to Work Program.

Research Funding Opportunities

Social Security Research Paper

View sample Social Security Research Paper. Browse other social sciences research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our research paper writing service for professional assistance. We offer high-quality assignments for reasonable rates.

‘Social Security’ denotes both a complex of public institutions to protect individuals against common life risks and the programmatic idea that a commonwealth should provide all of its members such protection that they may feel secure within that commonwealth. This concept is broader than ‘social insurance’ and narrower than ‘welfare state,’ but relates to similar problems.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% off with 24start discount code, 1. the emergence of the concept.

‘Security’ (Latin securitas) has a long-standing tradition in the political rhetoric in Europe since the Roman Empire. In early modern times it was often used together with ‘welfare’ and ‘felicity’ in order to paraphrase the duties of the prince to care for his subjects. Thomas Hobbes made ‘the safety of the people’ a central task for his Le iathan, and this idea became seminal for modern political theory and constitutional practice. Wilhelm von Humboldt (1792) was the first to distinguish clearly between ‘safety security’ and ‘welfare’: he restricted the task of the state to the provision of ‘safety security,’ i.e. on the one hand the protection of citizens’ freedom from external threats, and on the other hand the internal protection of citizens’ commitments and rights via the judiciary and police. ‘Welfare’ by contrast was now assigned to the private realm as the individual pursuit of happiness from which the state was to abstain from interfering. In the nineteenth century the idea of public safety and security became restricted to the defense and protection of individual rights, the latter becoming differentiated between security within the public realm and the reliability of legal protection. But in the twentieth century the idea of security became paramount and was related to psychology and psychoanalysis, to the theory of cognition, to industrial and technical safety as well as to social security.

The specific context for the emergence of the term ‘Social Security’ was the United States and the Great Depression. In American pragmatism ‘the quest for certainty’ was abandoned and superseded by ‘the search for security’ (John Dewey). Inspired by the ‘four wishes’ of William I Thomas, the ‘wish for security’ became a basic assumption about human motives and ‘insecurity’ a commonplace attribute of the Zeitgeist.

The invention of the expression ‘Social Security’ is commonly attributed to Franklin Delano Roosevelt or to his entourage. Roosevelt used it first in a Message to Congress on 30 September 1934 after having created a ‘Committee on Economic Security’ to prepare the ‘Social Security Act’ on 29 June 1934. However, a voluntary association promoting ‘Old Age Security’ renamed itself ‘American Association for Social Security’ in 1933 and changed the title of its journal from Old Age Security Herald to Social Security because the program of this association had been expanded from old-age pensions to all branches of social insurance (Rubinow 1934, pp. iii, 279).

In any case the expression quickly gained steam: it coined first the US ‘Social Security Act’ of 1935 (SSA). It then appeared in the Atlantic Charter of 1941, became the catchword for the Declaration of Philadelphia (1944) of the International Labour Organization (ILO) and found its way in 1948 into the Universal Declaration on Human Rights of the United Nations (Art. 22). After the Second World War ‘Social Security’ had become a leading term internationally in the programs for the development of welfare states. Its specific meaning remained vague, however.

2. Institutional Developments

At the time when the SSA was drafted, a public commitment to income maintenance for people in need was by no means a new idea. At the end of the nineteenth century two models were competing in Europe: public subsidizing of mutual benefit associations (called the ‘System of Gent’), and social insurance along the lines of the German (‘Bismarck’) Model. The British social reforms brought two new principles: payment of old-age pensions for deserving poors from the budget (1908), and flat-rate contributions and benefits in unemployment insurance (1911). In 1913 Sweden created the first universal system for securing income in old age. After the Russian Revolution the Bolsheviks drafted a comprehensive system of social protection against all major contingencies of life, following a concept which Lenin had already propagated in 1912. Starting in France (1932), hitherto voluntary child allowances of employers were made obligatory, and this ushered in a new model of administration: calling upon private bodies for the attainment of public social policy ends. Thus social insurance for dependent workers was already competing with other models of social protection. The victory of the term ‘social security’ over ‘social insurance’ was a consequence of its vague content which allowed the inclusion of all kinds of public social protection. In 1947 the ‘International Social Insurance Conference’ changed its name to ‘International Social Security Association’ and opened membership to state-administered systems (such as those of Britain and the US) which until then had been excluded.

Thus, the American SSA brought a new name but hardly new ideas. It had a problem, however, that was not present in Europe, namely federalism. The Supreme Court had long held attempts by the Federal Government to introduce measures of social protection to be unconstitutional. And it was only after a positive ruling by the Supreme Court on the SSA in 1937 that federal social policies took shape, including policies to combat the risks of the death of a breadwinner and of disability. Besides a federal system of Old-Age Insurance the SSA introduced a new federal instrument into American politics: matching grants to the states in order to induce them to abolish the old forms of the poor law, to introduce or to improve a system of unemployment compensation, and to provide better care to the blind as well as to mothers and children. Thus the hitherto strictly separated domains of the federal and state government became linked for the first time. This remained a continual subject of controversy in the American polity with lasting consequences for the structure of social protection: there exists a two-tiered system consisting of ‘social security’ on the federal and of ‘welfare’ on the state level. The first program, protecting most of the employed population, is well administered, highly stable, and positively valued. The second, providing help as a last resort, has remained an ever-contested object of repeated reforms. It is unevenly administered at the local level and its recipients remain stigmatized.

That which was innovative in the concept of social security was developed first in a report to the British Government entitled ‘Social Insurance and Allied Services’ in 1942, by a commission chaired by William Beveridge. It is the program of a comprehensi e and unified system, covering the whole population against all basic risks of life—an idea that Beveridge had already formulated in 1924. The Beveridge report met with unprecedented enthusiasm in the British population and became the blueprint for social legislation after the war.

The Beveridge report proved seminal also for the creation of the French ‘Securite Sociale’ (1946 48). But contrary to the case in the United Kingdom, the principle of administrative unification of the system met with strong resistance in France from various social groups which preferred to keep or create separate insurances meeting their specific needs and interests. In Germany only the term was received without changing the structure of the existing system. Moreover, a semantic distinction was made between the institutional (Soziale Sicherung) and the normative (Soziale Sicherheit) aspect of social security. In Scandinavia the term ‘Social Security’ was hardly received although the systems of social protection underwent substantial improvements.

Although many industrialized countries had developed measures of social protection against certain risks before World War II, their impact remained generally modest. Some groups, e.g. public servants or veterans, were privileged by public budgets, whereas industrial workers received only modest benefits out of contributions. With the exception of Scandinavia, the rural population remained mostly outside of existing systems. In most countries there existed a multitude of funds respectively designed for specific risks, groups, and localities. Repeated economic crises culminating in the Great Depression often impaired the reliability of expected benefits.

After the Second World War the new ideal of ‘Social Security’ was received differently in different countries. Almost everywhere coverage provided by existing institutions was extended or new institutions were created to protect all or most of the population against certain risks. Most countries assimilated the regulatory frameworks of different systems which covered the same risks, or integrated the systems. Finally, the risks covered by schemes underwent an international standardization. The Convention No. 102 of the ILO on ‘Minimum Standards of Social Security’ (1951) distinguishes nine forms of benefit: old-age benefits, survivor benefits, disability benefits, family allowances, unemployment benefits, sickness benefits, medical care, and maternity benefits. To be sure, most advanced industrial countries’ social protection institutions do not correspond one-for-one with that list of benefits, but today they indeed do tend to cover all of these risks, albeit with varying degrees of coverage and benefits.

From a structural perspective the institutions of Social Security form a delimited, though not everywhere equally defined system. Issues of delimitation concern the forms of benefit (in cash or in kind), of administration (centralized or decentralized; public, parapublic, or private), and of entitlement (public provision, insurance, relief), as well as the groups included (privileged or underprivileged groups are sometimes excluded) and the risks covered (family allowances, for example, are often excluded from the concept).

3. Security And Social Security As Value-Concepts

The administrative codification of Social Security as a complex of institutions which provide income replacement or benefits in kind should not obscure the fact that the success of the concept is due less to these institutions than to the value connotations of the concept. President Roosevelt evinced a keen sense of these connotations when he characterized social security as ‘freedom from want’ (6 January 1941), and when he described social security as a functional equivalent of earlier security deriving from ‘the interdependence of members of families upon each other and of the families within a small community upon each other’ (8 June 1934).

But what does security mean? And why did the concept become so prominent and pervasive in the most advanced industrial countries whose populations were more remote from serious want than any generations heretofore? Security relates to human attitudes towards the future and not the present. Insecurity does not mean an imminent danger but the possibility of future damages whose probability remains uncertain. The quest for security is a correlate to the growing complexity of society and the ensuing acceleration of social change. Thus security means not only protection, but also a subjective perception of the reliability of protection and the ensuing feeling of peace of mind. But in this psychological sense, security as self-assuredness can also mean the subjective capacity to bear the insecurity of risk. Hence on the individual level the concept of security remains highly ambivalent, and has indeed been abandoned by the discipline of psychology.

Security was also a longstanding pillar of political order which had already been incorporated into many constitutional documents and became transposed by Roosevelt into a new economic and social context. Just as government had the duty to care for external and internal security it was now to provide a framework to guarantee the means of existence for all, through full employment, social insurance, or other forms of benefits or services. ‘Social Security’ thus became the umbrella word for economic, social, and cultural rights as expressed in Article 22 of the United Nations Universal Declaration of Human Rights: ‘Everyone as a member of society, has the right to social security and is entitled to realization through national effort and international co-operation and in accordance with the organization and resources of each state, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality.’ These rights are then specified as the right to work (Art. 23), the right to recreation from work (Art. 24), the right to health, well-being, and social protection, with special emphasis on mothers and children (Art. 25), the right to education (Art. 26), and the right to participation in culture, science, and the arts (Art. 27).

Thus Social Security became another name for welfare in its political-philosophical sense. It became a key term in the international evolution of the welfare state after World War II. It was adopted differently on the national level, however. In most countries the institutional aspect has long dominated. In the actual context of ‘crisis‘ or ‘reconstruction’ of the welfare state ‘social security’ shows again its normative aspect: to what extent are politics free to dispose about political regulations which aim at stabilizing the future economic protection of the population or defined parts of it? ‘Social Security’ means the right of everybody who belongs to a certain commonwealth to be included into the basic provisions against destitution and want. And it suggests moreover that such provisions should take a coherent and clearly accessible form for everyone, so that people are not only legally protected but also have easy and reliable access to such protection.

4. Controversial Issues

4.1 general problems.

From the perspective of the Universal Declaration of Human Rights as well as William Beveridge or Pierre Laroque, the ‘father of the French securite sociale,’ Social Security was not restricted to publicly organized redistribution within populations at risk but also included policies of full employment and minimum wages. There was a clear awareness of the trade-off between full employment and the possibilities for sufficiently funding social protection. Social security as the goal of guaranteeing to the entire population the opportunity to participate in the economic and cultural life of its society implies education and employment as primary means and compensatory social protection only as a ‘second-best’ solution. The reification of ‘Social Security’ has obscured this original perspective and focuses public attention on processes of redistribution only. The current trend to substitute ‘work for welfare’ in the case of the able-bodied is quite in line with the original intention, provided that a decent minimum living standard is attainable at any rate. There is a strong divergence, however, with regard to whether the role of labor markets should be exclusive (as in the Anglo-Saxon world) or whether there should be a subsidiary public role (as in the Scandinavian and German-speaking countries) in employing the less productive parts of the active population.

From the perspective of the ‘Chicago School’ of economics, the existing systems of Social Security are overly costly, they disguise the relationship between an individual’s contributions and benefits, and they benefit the middle classes more than those who are truly in need. A negative income tax would better serve the poor, they maintain. Such a ‘rationalist’ economic perspective does not take into account the existing trust in public systems of reciprocal security in most countries. The ‘deal’ which pools the risks of the whole population in one scheme and thus makes basic economic security a matter of democratic politics seems to find high acceptance in most countries, although many people are aware of the likelihood of being among the net payers to the system.

Given the natural inequality of abilities and the social inequalites resulting from birth, education, labor markets, and competition in general, all political endeavors to foster social security—be it in the wider or in the narrower sense—face the problem of distributive effects. These relate not only to the amount of individual benefit and the forms of public financing but to the institutional arrangements as well. Strong political conflicts emerged in many countries around the issue of creating a single comprehensive system for certain provisions or for maintaining a fragmented system differentiating among classes of populations at risk. A somewhat related problem concerns the issue of securing only minimal standards or advanced standards of benefits through public provision. Finally, there is an obvious political struggle in all countries concerning the level of redistribution and who should bear the costs of social protection.

Comparative research has focused mainly on the explanation of national differences in institutional design and distributive effects. Path dependency upon early national approaches seems to be an important explanatory factor for these variations, beside political power relations and cultural orientations. An interesting finding is that systems which provide high levels of protection tend to protect minimal standards better as well.

4.2 Sectoral Aspects

Most political controversies do not arise with regard to the system of Social Security as a whole but concerning particular issues. As the functional organization of social protection is different from country to country, it remains difficult to generalize about this. One may identify some broad issues, however.

The paramount problem in the contemporary debate—particularly as a consequence of demographic aging—is the security of income in old age. A three-tiered system consisting of universal and stateadministered protection of basic standards (first tier), employment-related (and often private) protection of advanced standards (second tier), and complementary forms of personal provision (third tier) seems to be a promising model to resist the demographic and economic challenges of the decades to come. In order to ensure income security in old age for all, some public regulation and supervision of the second and third tiers remains important, however.

The second complex of risks is related to illness, industrial diseases, and long-term disability. Here, benefits in cash and in kind are necessary, and coverage for these risks is organized quite differently in different countries. Essentially, there are two different institutional models: the National Health Service and protection through social insurance. Both systems face the problem of containing the explosion of costs which has resulted from the compounding of a wide range of medical, technological, demographic, and economic factors.

The third complex of Social Security is related to the protection of the family. Since child labor has been forbidden and education made compulsory, children are no longer an economic asset for their parents, but rather a strong source of social inequality. It is still contested to what extent parents and especially mothers should be supported by public means, beyond the generally accepted subsidy in cases of manifest poverty. The low birth rate in most European populations now gives more political weight to demands to improve the living conditions and social protection of persons who are raising children or nursing their permanently incapacitated parents.

The most contested part of Social Security is public pro ision for the unemployed. Here, the trade-offs with labor-market policies and full employment are obvious, and convictions vary about how to deal with this problem.

The issue of poverty cuts across such functional distinctions. The extent to which it emerges as a separate problem which must be addressed by specific measures depends to a large extent on the institutional approaches which a country has adopted in the abovementioned realms of social security.

The issues discussed here are those generally attributed to the concept of social security. The concept of the welfare state covers additional services such as housing, education, and personal services.

Bibliography:

• Alber J 1982 Vom Armenhaus zum Wohlfahrtsstaat. Analysen zur Entwicklung on Sozial ersicherung in Westeuropa (From the Poor-House to the Welfare State. The Emergence of Social Insurance in Western Europe). Campus, Frankfurt-am-Main, Germany
• Arnold D R, Graetz M J, Munell A H (eds.) 1998 Framing the Social Security Debate. Values, Politics, and Economics. National Academy of Social Insurance, Washington, DC
• Atkinson A B 1989 Poverty and Social Security. Harvester Wheatsheaf, New York
• Baldwin S, Falkingham J (eds.) 1994 Social Security and Social Change. New Challenges to the Beveridge Model. Harvester Wheatsheaf, New York
• Burns E M 1956 Social Security and Public Policy. McGrawHill, New York
• Cohen W J, Friedman M 1972 Social Security: Universal or Selective? American Enterprise Institute, Washington, DC Committee on Economic Security 1937 Social Security in
• The Factual Background of the Social Security Act as Summarized from Staff Reports to the Committee on Economic Security. Published by the Social Security Board. Government Printing Office, Washington, DC
• Galant H 1955 Histoire politique de la securite sociale francaise 1945–1952 (A Political History of French Social Security 1945–1952). Armand Colin, Paris
• George V 1968 Social Security: Beveridge and After. Routledge & Kegan Paul, London
• Heclo H 1974 Modern Social Politics in Britain and Sweden: From Relief to Income Maintenance. Yale University Press, New Haven, CT
• Kaufmann F-X 1973 Sicherheit als soziologisches und sozialpolitisches Problem. Untersuchungen zu einer Wertidee hochdifferenzierter Gesellschaften (Security as a Problem for Sociology and Social Policy. An Inquiry into a Universal Value of Differentiated Societies), 2nd edn. Enke Lucius u. Lucius, Stuttgart, Germany
• Lowe R 1993 The Welfare State in Britain Since 1945. Macmillan, London
• Noble C 1997 Welfare As We Knew It. A Political History of the American Welfare State. Oxford University Press, New York
• Robbers G 1987 Sicherheit vals Menschenrecht (Security as a Human Right). Nomo, Baden-Baden, Germany
• Rubinow I M 1934 The Quest for Security. Henry Holt, New York
• Social Insurance and Allied Services 1942 Report by Sir William Beveridge. HMSO, London
• Weir M, Orloff A S, Skocpol T (eds.) 1988 The Politics of Social Policy in the United States. Princeton University Press, Princeton, NJ

ORDER HIGH QUALITY CUSTOM PAPER