2021 OASDI Trustees Report

skip to main content
Table of Contents Previous Next Tables Figures Index

 
B. LONG-RANGE ESTIMATES
The Trustees use three types of financial measures to assess the actuarial status of the Social Security trust funds under the financing approach specified in current law: (1) annual cash-flow measures, including income rates, cost rates, and balances; (2) trust fund ratios; and (3) summary measures such as actuarial balances and unfunded obligations.
The difference between the annual income rate and annual cost rate, both expressed as percentages of taxable payroll, is the annual balance. The level and trend of the annual balances at the end of the 75-year projection period are factors that the Trustees use to assess the financial condition of the program.
The trust fund ratio for a year is the proportion of the year’s projected cost that could be paid with fund reserves available at the beginning of the year. Critical factors considered by the Trustees in assessing actuarial status include: (1) the level and year of maximum trust fund ratio, (2) the year of depletion of the fund reserves and the percent of scheduled benefits that is still payable after reserves are depleted, and (3) the stability of the trust fund ratio at the end of the long-range period.
Solvency at any point in time requires that sufficient financial resources are available to pay all scheduled benefits at that time. Solvency is generally indicated by a positive trust fund ratio. “Sustainable solvency” for the financing of the program under a specified set of assumptions is achieved when the projected trust fund ratio is positive throughout the 75‑year projection period and is either stable or rising at the end of the period.
The Trustees summarize the total income and cost over valuation periods that extend through 75 years and over the infinite horizon.1 This section presents several summarized measures, including the actuarial balance and the open- group unfunded obligation. The actuarial balance indicates the size of any surplus or shortfall as a percentage of the taxable payroll over the period. The open-group unfunded obligation indicates the size of any shortfall in present-value dollars.
This section also includes additional information that the Trustees use to assess the financial status of the Social Security program, including: (1) a comparison of the number of beneficiaries to the number of covered workers, (2) the test of long-range close actuarial balance, and (3) the reasons for the change in the actuarial balance from the last report.
1. Annual Income Rates, Cost Rates, and Balances
The concepts of income rate and cost rate, expressed as percentages of taxable payroll, are important in the consideration of the long-range actuarial status of the trust funds. The annual income rate is the ratio of all non-interest income to the OASDI taxable payroll for the year. Non-interest income includes payroll taxes, taxes on scheduled benefits, and any General Fund transfers or reimbursements. The OASDI taxable payroll consists of the total earnings subject to OASDI taxes with some relatively small adjustments.2 The annual cost rate is the ratio of the cost of the program to the taxable payroll for the year. The cost includes scheduled benefits, administrative expenses, net interchange with the Railroad Retirement program, and payments for vocational rehabilitation services for disabled beneficiaries. For any year, the annual income rate minus the annual cost rate is the annual “balance” for the year.
Table IV.B1 presents a comparison of the estimated annual income rates and cost rates by trust fund and alternative. Table IV.B2 shows the separate components of the annual income rates.
Under the intermediate assumptions, the Trustees project that the OASI income rate will decrease from 11.62 percent of payroll for 2020 to 10.58 percent of payroll for 2021. The income rate in 2020 is relatively high because the numerator (credited payroll tax contributions in 2020) is based on estimates that did not anticipate the COVID-19 pandemic and ensuing recession, while the denominator (2020 taxable payroll) does reflect the actual lower-than-expected experience. The income rate then decreases in 2021 because of a negative adjustment to payroll tax contributions made in June 2021 reflecting the excess amounts credited in 2020. The OASI income rate rises to 11.11 percent in 2022 and then generally gradually rises thereafter, reaching 11.52 percent of taxable payroll for 2095. Income from taxation of benefits causes this gradual increase in the OASI income rate for two main reasons: (1) total scheduled benefits are rising faster than payroll; and (2) the ratio of total income tax on benefits to total benefits increases over time for reasons discussed in detail on page 154.
The OASI cost rate declines from 12.47 percent of payroll for 2020 to 12.33 percent for 20213 due primarily to a relatively large increase in taxable payroll in 2021, with 2020 taxable payroll being lower than expected because of the COVID-19 pandemic and ensuing recession. From 2021 to 2038, the OASI cost rate rises rapidly because the retirement of the baby-boom generation will continue to increase the number of beneficiaries much faster than the number of workers increases, as subsequent lower-birth-rate generations replace the baby-boom generation at working ages. From 2041 to 2046, the cost rate declines because the aging baby-boom generation is gradually replaced at retirement ages by the lower-birth-rate generations that followed. The OASI cost rate then rises from 15.10 percent in 2046 to 16.45 percent in 2078, largely because of the period of reduced birth rates starting with the recession of 2007-09, and then declines to 15.70 percent in 2095.
Projections of income rates under the low-cost and high-cost sets of assumptions are similar to those projected for the intermediate assumptions, because income rates are largely a reflection of the payroll tax rates specified in the law, with the changes from taxation of benefits noted above. In contrast, OASI cost rates for the low-cost and high-cost assumptions are significantly different from those projected for the intermediate assumptions. For the low-cost assumptions, the OASI cost rate generally rises from 12.03 percent in 2021 until it peaks in 2034 at 12.87 percent of payroll. The cost rate then declines to 12.02 percent for 2055, rises to 12.11 percent for 2068, and then mostly declines to 10.97 percent in 2095, at which point the income rate reaches 11.23 percent. For the high-cost assumptions, the OASI cost rate rises through 2087 and then declines gradually through 2095. The cost rate rises relatively rapidly through about 2038 because of the aging of the baby-boom generation. The OASI cost rate reaches 24.67 percent of payroll for 2095, at which point the income rate reaches 12.06 percent.
The pattern of the projected OASI annual balance is important in the analysis of the financial condition of the program. Under the intermediate assumptions, the annual balance is negative throughout the projection period. The annual deficit increases significantly in 2021 because of the adjustment for lower-than-expected payroll tax revenue in 2020 mentioned earlier. After 2022, the annual deficit mostly rises to 3.72 percent for 2041. It then declines to 3.65 percent of payroll for 2046, rises to 4.89 percent for 2078, and declines thereafter, reaching 4.18 percent of taxable payroll for 2095.
Under the low-cost assumptions, the OASI annual deficit decreases from 1.54 percent of payroll for 2021 to 0.84 percent in 2022, and then rises to 1.58 percent of payroll for 2034. After 2034, the annual deficit declines to 0.74 percent of payroll for 2055, and rises to 0.81 percent for 2067. After 2067, the OASI annual balance generally improves, turning positive in 2088, and reaching 0.26 percent of payroll for 2095. Under the high-cost assumptions, the OASI annual deficit rises from 1.97 percent for 2021 to 12.87 percent for 2087, then declines relatively modestly to 12.62 percent for 2095.
Income
rate a
Cost
rateb

a
Income rates include certain reimbursements from the General Fund of the Treasury.

b
Benefit payments scheduled to be paid on January 3 are actually paid on December 31 as required by the statutory provision for early delivery of benefit payments when the normal payment delivery date is a Saturday, Sunday, or legal public holiday. For comparability with the values for historical years and the projections in this report, all trust fund operations and asset reserves reflect the 12 months of benefits scheduled for payment each year.

c
Between -0.005 and 0 percent of taxable payroll.

d
The annual balance is projected to be negative for a temporary period and return to positive levels before the end of the projection period.

Notes:
1. The income rate excludes interest income.
2. Revisions of taxable payroll may change some historical values.
3. Components may not sum to totals because of rounding.
Under the intermediate assumptions, the projected DI cost rate declines from 1.90 percent for 2020 to 1.79 percent for 2021 due primarily to a relatively large increase in taxable payroll for 2021, with 2020 taxable payroll being lower than expected because of the COVID-19 pandemic and ensuing recession. After 2022, the cost rate generally declines more gradually to 1.73 percent for 2031. Then the DI cost rate increases gradually to 2.03 percent for 2055. Thereafter, the cost rate remains relatively stable, decreasing slowly to 1.91 percent for 2083, and then increasing to 2.00 percent for 2095. Except for the decline in payroll tax contributions in 2021 from the adjustment reflecting lower-than-expected 2020 payroll tax contributions, the projected DI income rate is relatively stable throughout the projection period, reaching 1.84 percent for 2095. The annual balance is negative in 2021 and turns positive in 2022, reaching a peak of 0.10 percent of payroll for 2031. The annual balance then declines and becomes negative in 2040. After 2040, the DI annual deficit increases to 0.19 percent in 2055, generally decreases to 0.08 percent in 2083, and then increases to 0.17 percent of payroll for 2095.
Under the low-cost assumptions, the projected DI cost rate declines from 1.74 percent of payroll for 2021 to 1.29 percent for 2037, and increases to 1.37 percent for 2052. The cost rate then declines through 2081 and increases slowly thereafter, reaching 1.34 percent for 2095. The annual balance is negative for 2021 and positive throughout the remainder of the long-range period. Under the high-cost assumptions, the DI cost rate rises from 1.82 percent of payroll in 2021 to 2.93 percent in 2056 and fluctuates thereafter, reaching 2.86 percent for 2095. The annual balance is negative throughout the projection period, with annual deficits rising from 0.10 percent of payroll for 2021 to 1.08 percent in 2056, and attaining 1.01 percent for 2095.
Figure IV.B1 shows the patterns of the historical and projected OASI and DI annual cost rates. The patterns in projected OASI and DI cost rates are described earlier in this chapter. Historical annual OASI cost rates shifted upward starting in 2008 and have remained at relatively high levels since then, primarily due to the retirement of the baby-boom generation. Historical annual DI cost rates rose substantially between 1990 and 2010 in large part due to: (1) aging of the working population as the baby-boom generation moved from ages 25-44 in 1990, where disability prevalence is low, to ages 45-64 in 2010, where disability prevalence is much higher; (2) a substantial increase in the percentage of women insured for DI benefits as a result of increased and more consistent rates of employment; and (3) increased disability incidence rates for women to a level similar to those for men by 2010. As of 2010, these three factors have largely stabilized. Other factors that are not yet fully understood, including the changing nature of work, have caused age-sex-adjusted incidence rates and cost rates to decline from 2010 to 2020. Figure IV.B1 shows only the income rates for alternative II because the variation in income rates by alternative is very small. Income rates generally increase slowly for each of the alternatives over the long-range period. Taxation of benefits, which is a small portion of income, is the main source of the increases in the income rate and the variation among the alternatives.
Table IV.B1 shows the annual balances for OASI, DI, and OASDI. The pattern of the annual balances is important to the analysis of the financial condition of the Social Security program as a whole. As seen in figure  IV.B1, the magnitude of each of the positive balances is the distance between the appropriate cost-rate curve and the income-rate curve above it. The magnitude of each of the deficits is the distance between the appropriate cost-rate curve and the income-rate curve below it. Annual balances follow closely the pattern of annual cost rates after 1990 because the payroll tax rate does not change for the OASDI program, with only small variations in the allocation between DI and OASI except for changes due to the 1994 and the 2016 through 2018 payroll tax rate reallocations.
In the future, the costs of OASI, DI, and the combined OASDI programs as a percentage of taxable payroll are unlikely to fall outside the range encompassed by alternatives I and III because alternatives I and III define a wide range of demographic and economic conditions.
Long-range OASDI cost and income are most often expressed as percentages of taxable payroll. However, the Trustees also present cost and income as shares of gross domestic product (GDP), the value of goods and services produced during the year in the United States. Under alternative II, the Trustees project OASDI cost to increase from about 5.1 percent of GDP for 2021 to about 6.0 percent for 2039. After 2039, OASDI cost as a percentage of GDP declines slightly through 2051, increases to a peak of 6.2 percent for 2077, and thereafter decreases slowly, reaching about 5.9 percent by 2095. Appendix G presents full estimates of income and cost relative to GDP.
Table IV.B2 contains historical and projected annual income rates and their components by trust fund and alternative. The annual income rates consist of the scheduled payroll tax rates, the rates of income from taxation of scheduled benefits, and the rates of income from General Fund reimbursements. Projected income from taxation of benefits increases over time for reasons discussed on page 154.
Taxation of benefitsa
General Fund Reimbursementsb
Totalc

a
Revenue from taxation of benefits is the amount that would be assessed on benefit amounts scheduled in the law.

b
Includes payroll tax revenue forgone under the provisions of Public Laws 111-147, 111-312, 112-78, and 112‑96, and other miscellaneous reimbursements.

c
Values exclude interest income.

d
Between 0 and 0.005 percent of taxable payroll.

Note: Components may not sum to totals because of rounding.
2. Comparison of Workers to Beneficiaries
Under the intermediate assumptions, the Trustees project that the OASDI cost rate will rise rapidly between 2021 and 2038, primarily because the number of beneficiaries rises much more rapidly than the number of covered workers as the baby-boom generation retires. The ratio of OASDI beneficiaries to workers is dominated by the OASI program because all workers eventually die or retire, but only a relatively small minority become disabled. The trends described below are primarily due to demographic changes and thus affect the DI program roughly 20 years earlier than the OASI and OASDI programs. The baby-boom generation had lower fertility rates than their parents, and the Trustees expect that lower fertility rates will persist for all future generations; therefore, the ratio of OASDI beneficiaries to workers will rise rapidly and reach a permanently higher level after all of the baby-boom generation has retired. Due to increasing longevity, the ratio of beneficiaries to workers will generally rise slowly thereafter. Table IV.B3 provides a comparison of the numbers of covered workers and beneficiaries.
Covered
workers a
(in thousands)
Beneficiaries b (in thousands)
OASDIc

a
Workers who are paid at some time during the year for employment on which OASDI taxes are due.

b
Beneficiaries with monthly benefits in current-payment status as of June 30.

c
This column is the sum of OASI and DI beneficiaries. A small number of beneficiaries receive benefits from both funds.

Notes:
1. The number of beneficiaries does not include uninsured individuals who received benefits under section 228 of the Social Security Act. The General Fund of the Treasury reimbursed the trust funds for the costs of most of these individuals.
2. Historical covered worker and beneficiary data are subject to revision.
3. Components may not sum to totals because of rounding.
The effect of the demographic shift under the three alternatives on the OASDI cost rates is clear when one considers the projected number of OASDI beneficiaries per 100 covered workers. Compared to the 2020 level of 37 beneficiaries per 100 covered workers, the Trustees project that this ratio rises to 44 by 2038 under the intermediate assumptions because the growth in beneficiaries greatly exceeds the growth in workers. By 2095, this projected ratio rises further under the intermediate and high-cost assumptions, reaching 47 under the intermediate assumptions and 66 under the high-cost assumptions. Under the low-cost assumptions, this ratio rises to 41 by 2038 and then generally declines, reaching 36 by 2095. Figure IV.B2 shows beneficiaries per 100 covered workers.
For each alternative, the curve in figure IV.B2 is strikingly similar to the corresponding cost-rate curve in figure IV.B1. This similarity emphasizes the extent to which the cost rate is determined by the age distribution of the population. The cost rate is essentially the product of the number of beneficiaries and their average benefit, divided by the product of the number of covered workers and their average taxable earnings. For this reason, the pattern of the annual cost rates is similar to that of the annual ratios of beneficiaries to workers.
Table IV.B3 also shows the number of covered workers per OASDI beneficiary, which was about 2.7 for 2020. Under the intermediate assumptions, this ratio declines generally throughout the long-range period, reaching 2.3 for 2038 and 2.1 by 2095. Under the low-cost assumptions, this ratio declines to 2.4 for 2038, then generally rises to 2.8 by 2095. Under the high-cost assumptions, this ratio decreases steadily to 1.5 by 2095.
3. Trust Fund Ratios and Test of Long-Range Close Actuarial Balance
Trust fund ratios are critical indicators of the adequacy of the financial resources of the Social Security program. The trust fund ratio for a year is the amount of asset reserves in a fund at the beginning of a year expressed as a percentage of the cost for the year. Under present law, the OASI and DI Trust Funds do not have the authority to borrow other than in the form of advance tax transfers, which are limited to expected taxes for the current calendar month. If reserves held in either trust fund become depleted during a year, and continuing revenue falls short of the cost of scheduled benefits, then full scheduled benefits would not be payable on a timely basis. For this reason, the trust fund ratio is a critical financial measure.
The trust fund ratio serves an additional important purpose in assessing the actuarial status of the program. If the projected trust fund ratio is positive throughout the period and is either level or increasing at the end of the period, then projected adequacy for the long-range period is likely to continue for subsequent reports. Under these conditions, the program has achieved sustainable solvency.
Table IV.B4 shows the Trustees’ projections of trust fund ratios by alternative, without regard to advance tax transfers that would be effected, for the separate and combined OASI and DI Trust Funds. The table also shows the years of trust fund reserve depletion and the percentage of scheduled benefits that would be payable thereafter, by alternative.
Under the intermediate assumptions, the OASI trust fund ratio is projected to decline from 280 percent at the beginning of 20214 until the trust fund reserves become depleted in 2033 (one year earlier than projected in last year’s report), at which time 76 percent of scheduled benefits would be payable. The DI trust fund ratio decreases from 66 percent at the beginning of 2021 to 62 percent in 2022. After 2022, the DI trust fund ratio increases for a time, reflecting the period of recent and near-term expected low incidence rates, reaching a peak of 116 percent by 2040. The ratio then declines until the trust fund reserves become depleted in 2057 (8 years earlier than projected in last year’s report). At that time, 91 percent of scheduled benefits would be payable. The differences in depletion years for both OASI and DI from last year’s report are largely due to the revenue effects from the 2020 recession.
Under the intermediate assumptions, the trust fund ratio for the combined OASI and DI Trust Funds declines from 253 percent at the beginning of 2021 until the combined fund reserves become depleted early in 2034 (one year earlier than projected in last year’s report), at which time 78 percent of scheduled benefits would be payable.
Under the low-cost assumptions, the trust fund ratio for the DI program increases from 2022 throughout the remainder of the projection period, from 65 percent at the beginning of 2022 to the extremely high level of 2,981 percent for 2096. For the OASI program, the trust fund ratio declines steadily, from 280 percent for 2021 until the reserves become depleted in 2040, at which time 88 percent of scheduled benefits would be payable. For the combined OASDI program, the trust fund ratio declines from 253 percent for 2021 until the combined fund reserves become depleted in 2061. However, the trust funds would have sufficient income by the end of 2092 to permit full payment of scheduled benefits thereafter and also to pay in arrears the temporary shortfalls between 2061 and 2092. By 2093, trust fund reserves become positive and the trust fund ratio increases thereafter, to a ratio of 23 percent in 2096. Because the DI trust fund ratio is positive throughout the projection period and increasing at the end of the period, under the low-cost assumptions, only the DI program achieves sustainable solvency.
Under the high-cost assumptions, the OASI trust fund ratio declines from 280 percent for 2021 until reserves become depleted in 2031, at which time 68 percent of scheduled benefits would still be payable. The DI trust fund ratio declines from 66 percent for 2021 until the reserves become depleted in 2027. At that time, 87 percent of scheduled benefits would still be payable. The combined OASI and DI trust fund ratio declines from 252 percent for 2021 until reserves become depleted in 2031, at which time 70 percent of scheduled benefits would still be payable.
The Trustees project trust fund reserve depletion within the 75-year projection period with the exception of the DI Trust Fund under the low-cost assumptions. It is therefore very likely that lawmakers will need to increase income, reduce program costs, or both, in order to maintain solvency for the trust funds. The stochastic projections discussed in appendix E suggest that trust fund reserve depletion is highly probable before mid-century.
In the 2020 report, the Trustees projected that the combined trust fund reserves would become depleted in 2031 and 2035 under the high-cost and intermediate assumptions, respectively, and would become temporarily depleted between 2079 and 2088 under the low-cost assumptions.
Table IV.B4.—Trust Fund Ratios, Calendar Years 2021-2095a
b

a
Benefit payments scheduled to be paid on January 3 are actually paid on December 31 as required by the statutory provision for early delivery of benefit payments when the normal payment delivery date is a Saturday, Sunday, or legal public holiday. For comparability with the values for historical years and the projections in this report, all trust fund ratios reflect the 12 months of benefits scheduled for payment each year.

b
Trust fund reserves would be depleted at the beginning of this year.

c
Trust fund reserves would not be depleted within the projection period.

d
Trust fund reserves would be depleted for a temporary period, and return to positive levels before the end of the period.

Note: The definition of trust fund ratio appears in the Glossary. The ratios shown for the combined trust funds for years after reserve depletion of either the DI or OASI Trust Fund are hypothetical.
Since 2013, when the Trustees modified the test of long-range close actuarial balance, the standard for each trust fund requires meeting two conditions: (1) the test of short-range financial adequacy is satisfied; and (2) the trust fund ratios stay above zero throughout the 75-year projection period, allowing scheduled benefits to be paid in a timely manner throughout the period. Both the long-range test and the short-range test are applied based on the intermediate set of assumptions. As discussed in section IV.A, the DI Trust Fund fails the test of short-range financial adequacy because the trust fund ratio does not reach 100 percent during the 10-year period. The OASI and combined OASI and DI Trust Funds fail the test of short-range financial adequacy because the trust fund ratios drop below 100 percent by the end of the 10‑year period. Under the intermediate assumptions, the OASI Trust Fund reserves become depleted in 2033, DI Trust Fund reserves become depleted in 2057, and the combined OASI and DI Trust Fund reserves become depleted in 2034. Therefore, the OASI, DI, and combined OASI and DI Trust Funds all fail the test of long-range close actuarial balance.
Figure IV.B3 illustrates the trust fund ratios for the separate OASI and DI Trust Funds for each of the alternative sets of assumptions. DI Trust Fund status is more uncertain than OASI Trust Fund status because there is a high degree of uncertainty associated with future disability prevalence. A graph of the trust fund ratios for the combined trust funds appears in figure II.D6.
4. Summarized Income Rates, Summarized Cost Rates, and Actuarial Balances
Summarized values for the full 75-year period are useful in analyzing the program’s long-range financial adequacy over the period as a whole, both under present law and under proposed modifications to the law. All annual amounts included in a summarized value are present-value discounted to the valuation date. It is important to note that the actuarial balance indicates the solvency status of the fund only for the very end of the period.
Table IV.B5 presents summarized income rates, summarized cost rates, and actuarial balances for 25-year, 50-year, and 75-year valuation periods. Summarized income rates are the sum of the present value of non-interest income for a period (which includes scheduled payroll taxes, the projected income from the taxation of scheduled benefits, and reimbursements from the General Fund of the Treasury) and the starting trust fund asset reserves, expressed as a percentage of the present value of taxable payroll over the period. Under current law, the total OASDI payroll tax rate will remain at 12.4 percent in the future. In contrast, the Trustees expect income from taxation of benefits, expressed as a percentage of taxable payroll, to increase in most years of the long-range period for the reasons discussed on page 154. Summarized cost rates are the sum of the present value of cost for a period (which includes scheduled benefits, administrative expenses, net interchange with the Railroad Retirement program, and payments for vocational rehabilitation services for disabled beneficiaries) and the present value of the cost of reaching a target trust fund of 100 percent of annual cost at the end of the period, expressed as a percentage of the present value of taxable payroll over the period.
The actuarial balance for a valuation period is equal to the difference between the summarized income rate and the summarized cost rate for the period. An actuarial balance of zero for any period indicates that cost for the period could be met for the period as a whole (but not necessarily at all points within the period), with a remaining trust fund reserve at the end of the period equal to 100 percent of the following year’s cost. A negative actuarial balance for a period indicates that the present value of income to the program plus the existing trust fund is less than the present value of the cost of the program plus the cost of reaching a target trust fund reserve of one year’s cost by the end of the period. Generally, a trust fund is deemed to be adequately financed for a period if the actuarial balance is zero or positive, meaning that the reserves at the end of the period are at least equal to annual cost. Note that solvency is possible with a small negative actuarial balance where reserves are still positive.5
Table IV.B5 contains summarized rates for the intermediate, low-cost, and high-cost assumptions. The low-cost and high-cost assumptions define a wide range of possibilities. Financial outcomes as good as the low-cost scenario or as bad as the high-cost scenario are unlikely to occur.
For the 25-year valuation period, the OASDI program has an actuarial balance of -0.37 percent of taxable payroll under the low-cost assumptions, ‑2.37 percent under the intermediate assumptions, and -4.66 percent under the high-cost assumptions. These balances indicate that the program is not adequately financed for the 25‑year valuation period under any of these three sets of assumptions.
For the 50‑year valuation period, the OASDI program has actuarial balances of -0.31 percent under the low-cost assumptions, ‑3.14 percent under the intermediate assumptions, and ‑6.79 percent under the high-cost assumptions. These actuarial balances mean that the OASDI program is not adequately financed for the 50‑year valuation period under any of these three sets of assumptions.
For the entire 75-year valuation period, the combined OASDI program has actuarial balances of -0.12 percent of taxable payroll under the low-cost assumptions, ‑3.54 percent under the intermediate assumptions, and ‑8.25 percent under the high-cost assumptions. These balances indicate that the combined OASDI program is not adequately financed for the 75-year valuation period under any of these three sets of assumptions.
Assuming the intermediate assumptions accurately capture future demographic and economic trends, solvency for the program over the next 75 years could be restored using a variety of approaches. For example, revenue could be increased in a manner equivalent to an immediate and permanent increase in the combined Social Security payroll tax rate from 12.40 percent to 15.76 percent (a relative increase of 27.1 percent),6 cost could be reduced in a manner equivalent to an immediate and permanent reduction in scheduled benefits of about 21 percent, or some combination of approaches could be used.
However, eliminating the actuarial deficit for the next 75-year valuation period requires raising payroll taxes or lowering benefits by more than is required just to achieve solvency, because the actuarial deficit includes the cost of attaining a target trust fund equal to 100 percent of annual program cost by the end of the period. The actuarial deficit could be eliminated for the 75-year period by increasing revenue in a manner equivalent to an immediate and permanent increase in the combined payroll tax from 12.40 percent to 15.94 percent (a relative increase of 28.5 percent),7 reducing cost in a manner equivalent to an immediate reduction in scheduled benefits of about 22 percent, or some combination of approaches could be used.
Under the intermediate assumptions, the OASDI program has large annual deficits toward the end of the long-range period that reach 4.34 percent of payroll for 2095 (see table IV.B1). These large deficits indicate that annual cost continues to exceed non-interest income after 2095, so continued adequate financing would require larger changes than those needed to maintain solvency for the 75-year period. Over the period extending through the infinite horizon, the actuarial deficit is 4.6 percent of payroll under the intermediate assumptions.
Beginning
asset reservesa

a
Benefit payments scheduled to be paid on January 3 are actually paid on December 31 as required by the statutory provision for early delivery of benefit payments when the normal payment delivery date is a Saturday, Sunday, or legal public holiday. For comparability with the values for historical years and the projections in this report, all trust fund operations and asset reserves reflect the 12 months of benefits scheduled for payment each year.

b
Between -0.005 and 0 percent of taxable payroll.

Note: Components may not sum to totals because of rounding.
5. Open-Group Unfunded Obligation
Consistent with practice since 1965, this report focuses on a 75-year open-group valuation to evaluate the long-run financial status of the OASDI program. The open-group valuation includes non-interest income and cost for past, current, and future participants through the year 2095. The open-group unfunded obligation measures the adequacy of financing over the period as a whole for a program financed on a pay-as-you-go basis. On this basis, payroll taxes and scheduled benefits for all participants are included through 2095.
The open-group unfunded obligation increased from $16.8 trillion shown in last year's report to $19.8 trillion in this report. If there had been no changes in starting values, assumptions, laws, or methods for this report, then the open-group unfunded obligation would have increased to $17.5 trillion solely due to the change in the valuation period. This expected increase in the unfunded obligation occurs because: (1) the unfunded obligation is now discounted to January 1, 2021, rather than to January 1, 2020, which tends to increase the unfunded obligation by the annual nominal interest rate; and (2) the unfunded obligation now includes an additional year (2095). However, changes in the law, assumptions, methods, and starting values resulted in a net $2.3 trillion increase in the unfunded obligation.
The 75-year unfunded obligation is equivalent to 3.35 percent of OASDI taxable payroll and 1.2 percent of GDP for 2021-95.8 These percentages were 3.03 and 1.0, respectively, for last year’s report. The 75-year unfunded obligation as a percentage of taxable payroll is less than the actuarial deficit, because the unfunded obligation excludes the cost of having an ending target trust fund value.
The actuarial deficit was 3.21 percent of payroll in last year’s report, and was expected to increase to a deficit of 3.27 percent of payroll solely due to the change in the valuation period. Changes in the law, assumptions, methods, and starting values combined to account for an additional 0.27 percentage point increase (worsening) in the actuarial deficit to 3.54 percent of payroll. The actuarial deficit is 1.2 percent of GDP in this year’s report, increased from 1.1 percent in last year’s report.
As mentioned above, the open-group unfunded obligation expressed in dollars, as well as the actuarial deficit, increased (worsened) more than would be expected from changing the valuation period alone. This increase occurred for a variety of reasons described in the next section, primarily reductions in near-term interest rates and updates to methods and programmatic data.
Table IV.B6 presents the components and the calculation of the long-range (75-year) actuarial balance under the intermediate assumptions. The present value of future cost less future non-interest income over the long-range period, minus the amount of trust fund asset reserves at the beginning of the projection period, is $19.8 trillion for the OASDI program. This amount is the 75-year “open-group unfunded obligation” (see row H). The actuarial deficit (which is the negative of the actuarial balance) combines this unfunded obligation with the present value of the ending target trust fund and expresses the total as a percentage of the present value of the taxable payroll for the period. The present value of future non-interest income minus cost, plus starting trust fund reserves, minus the present value of the ending target trust fund, is ‑$20.9 trillion for the OASDI program.
E.

a
Less than $0.5 billion.

b
The calculation of the actuarial balance includes the cost of accumulating a target trust fund reserve equal to 100 percent of annual cost at the end of the period.

Note: Components may not sum to totals because of rounding.
Consideration of summary measures alone (such as the actuarial balance and open-group unfunded obligation) for a 75-year period can lead to incorrect perceptions and to policy prescriptions that do not achieve sustainable solvency. These concerns can be addressed by considering the trend in trust fund ratios toward the end of the period. (See the discussion of “sustainable solvency” beginning on page 53.)
Another measure of trust fund finances, discussed in appendix F, is the infinite horizon unfunded obligation, which takes account of all annual balances, even those after 75 years. The extension of the time period past 75 years assumes that the current-law OASDI program and the demographic and economic trends used for the 75‑year projection continue indefinitely. This infinite horizon unfunded obligation is estimated to be 4.6 percent of taxable payroll or 1.4 percent of GDP. These percentages were 4.6 and 1.4, respectively, for last year’s report. Of course, the degree of uncertainty associated with estimates increases substantially for years further in the future.
6. Reasons for Change in Actuarial Balance From Last Report
Table IV.B7 shows the effects of changes on the long-range actuarial balance under the intermediate assumptions, by category, between last year’s report and this report.
Valuation period a

a
The change in the 75-year valuation period from last year’s report to this report means that the 75-year actuarial balance now includes the relatively large negative annual balance for 2095. This change in the valuation period results in a larger long-range actuarial deficit. The actuarial deficit includes the trust fund reserve at the beginning of the projection period.

Note: Components may not sum to totals because of rounding.
If the law, data, assumptions, and methods had all remained unchanged from last year’s Trustees Report, the long-range OASDI actuarial balance would have decreased (become more negative) by 0.06 percent of taxable payroll solely due to the change in the valuation period. However, as described below, projections in this report also reflect new data and changes in law, assumptions, and methods. These changes, including the change in the valuation period, combine to decrease (worsen) the long-range OASDI actuarial balance by 0.32 percentage point, from -3.21 percent of taxable payroll in last year’s report to -3.54 percent in this report.
Legislation/Regulation
Changes in law, regulations, and policy result in a decrease in the long-range OASDI actuarial balance of 0.01 percent of payroll.
Since the last report, there has been one change in policy that is expected to have a significant financial effect on the long-range actuarial balance. The estimates presented in last year’s report reflected the assumption that the Deferred Action for Childhood Arrivals (DACA) policy would be phased out, consistent with the Department of Homeland Security’s September 5, 2017 rescission of the 2012 DACA policy directive. This year’s report incorporates the effects of two developments during 2020 and early 2021. First, pursuant to a United States District Court order, the US Citizenship and Immigration Services resumed accepting first-time applications and renewal requests for participation in the DACA program, effective December 7, 2020. Second, on January 20, 2021, President Biden issued a memorandum directing the Secretary of Homeland Security to take appropriate action, in consultation with the Attorney General, to preserve and fortify DACA consistent with applicable law. This change to preserve DACA extends indefinitely the ability of those qualifying to remain in the country and work lawfully. The effect over the next 75 years is to increase future benefits slightly more than future payroll tax revenue because: (1) a significant portion of the payroll taxes from this group has already been credited to the trust funds, while the vast majority of the OASDI benefits they will earn will be in the future, dependent on their preserving DACA status; and (2) currently scheduled payroll tax rates are not sufficient to fully finance future benefits for this group and in general. This change in policy is thus estimated to decrease the long-range actuarial balance by 0.01 percent of taxable payroll.
Valuation Period
As mentioned above, changing the 75-year valuation period from 2020 through 2094 for last year’s report to 2021 through 2095 for this report decreases the projected long-range OASDI actuarial balance by 0.06 percent of taxable payroll. This decrease is mainly the result of including the relatively large negative annual balance for 2095 in this year’s 75-year projection period. Note that the actuarial balance calculation includes trust fund reserves at the beginning of the projection period. These reserves reflect the program’s net financial flows for all past years, including 2020, up to the start of the valuation period.
Demographic Data and Assumptions
New demographic data and changes in demographic assumptions combine to increase (improve) the long-range OASDI actuarial balance by 0.07 percent of taxable payroll.
Two ultimate demographic assumptions were changed for this year’s report. First, future birth rates are now projected using a cohort-based model. (This change in methodology is described in detail and quantified separately in the “Methods and Programmatic Data” section below.) With this new cohort approach, the Trustees increased the ultimate total fertility rate from 1.95 to 2.00. This ultimate rate will not be realized on a period basis until 2056, which is about 25 years later than the ultimate rate was reached in last year’s report. This change in the ultimate fertility assumption, after making the change in method, increases the actuarial balance by 0.09 percent of taxable payroll. (As indicated later, this effect due to the change in assumption is fully offset by the effect of the change in method.)
Second, this year’s report adds an additional cause of death category by separating dementia out from the all-other-causes category, due to its very different historical trend and prospects for future improvement. At the same time, ultimate rates of mortality improvement have been updated for cardiovascular disease for all age groups and for the all-other-causes category at ages 85 and over. These changes to the ultimate mortality assumptions increase the actuarial balance by 0.04 percent of taxable payroll.
Updates to near-term mortality assumptions to reflect the effects of the COVID-19 pandemic increase the long-range actuarial balance by 0.04 percent of taxable payroll. Death rates were increased significantly for 2020 and 2021, and to a lesser extent for 2022 and 2023, to account for the elevated deaths during the pandemic period. Projected death rates for years after 2023 are unchanged from the levels that would have been projected in the absence of the pandemic.
Several demographic data updates also resulted in significant changes in the long-range actuarial balance. First, birth data through the third quarter of 2020 indicate somewhat lower fertility rates than were assumed in last year’s report for 2020. These updated data result in lower birth rates during the transition period to the ultimate level, decreasing the actuarial balance by 0.04 percent of taxable payroll. Second, new mortality data (final 2018 and 2019 data for ages under 65 from NCHS; final 2017 and preliminary 2018 and 2019 data for ages 65 and older from Medicare experience) were incorporated into the projection. Without considering the other changes to death rates, these new data alone result in slightly lower death rates for all future years, decreasing the actuarial balance by 0.01 percent of taxable payroll. Third, updates to lawful permanent resident (LPR) immigration data decrease the actuarial balance by 0.01 percent of taxable payroll. Finally, updates to historical population data, other-than-LPR immigration data, marriage and divorce data, and other minor data updates combine to decrease the actuarial balance by 0.03 percent of taxable payroll.
Economic Data and Assumptions
New economic data and changes in economic assumptions, in combination, have a negligible net effect on the long-range OASDI actuarial balance.
Two ultimate economic assumptions were changed for this year’s report. First, the ultimate age-sex-adjusted unemployment rate is reduced from 5.0 percent for last year’s report to 4.5 percent for this year’s report. At the same time, the labor force projection model was updated to incorporate data from the latest complete economic cycle (2007-19), and to reflect changes in the relationships among factors affecting labor force participation and employment. (This change in methodology is described in more detail and quantified separately in the “Methods and Programmatic Data” section below.) As mentioned in last year’s report, the Trustees believe that a structural shift has occurred in relationships between employment, unemployment, and the labor force, and thus future long-term unemployment rates and labor force participation rates will both remain somewhat below long-term past averages. This change to the ultimate unemployment rate assumption alone increases the actuarial balance by 0.07 percent of taxable payroll.
Second, the average real wage differential over the last 65 years of the projection period for this year’s report is 1.15 percentage points per year, slightly higher than the 1.14 value in last year’s report. The real wage differential assumptions for the first ten years of the projection period were also increased. These higher long-term and near-term real wage differential assumptions are based primarily on updated projections by the Centers for Medicare and Medicaid Services (CMS) of slower assumed growth in employer-sponsored group health insurance premiums. Because these premiums are not subject to the payroll tax, slower growth in these premiums means that a larger share of employee compensation will be in the form of wages that are subject to the payroll tax. These changes to the real wage differential assumptions increase the long-range actuarial balance by 0.04 percent of taxable payroll.
Although the ultimate real interest rate assumption remains at 2.3 percent for this year’s report, interest rates over the next ten years have been adjusted downward significantly. Real interest rates are now assumed to be negative for calendar years 2021 through 2024, with a gradual rise to the ultimate real interest rate after the economy has fully recovered from the recession. This change to the near-term real interest rate assumptions decreases the actuarial balance by 0.03 percent of payroll.
Other changes to historical data and near-term economic assumptions combine for a net decrease in the actuarial balance of 0.09 percent of taxable payroll. The pandemic and ensuing recession have had dramatic effects on economic experience during 2020 and early 2021, and, in turn, the expected path of the economy over the next few years. In particular, the level of potential GDP is assumed to be roughly 1 percent lower than the level in last year’s report beginning with the second quarter of 2020. This shift to a lower level of potential GDP reflects the expectation of a permanent level shift due to the effects of the recession on total economy labor productivity and the size of the population at ages 16 and over.
Disability Data and Assumptions
New disability data and changes in near-term disability assumptions have a negligible net effect on the long-range OASDI actuarial balance.
Methods and Programmatic Data
The projections in this report also reflect several methodological improvements and updates based on new program-specific data. These methodological changes, programmatic data updates, and interactions combine to decrease the long-range OASDI actuarial balance by 0.33 percent of taxable payroll. Descriptions of six significant methodological changes and programmatic data updates follow.
First, as mentioned earlier under the “Demographic Data and Assumptions” section, future birth rates are now projected using a cohort-based approach, rather than the period-based approach used in the 2020 and prior reports. The cohort method provides a direct reflection of the shift in childbearing to older ages that has been observed since the end of the baby-boom period in 1965, and results in a much more extended transition to ultimate birth rates from the current low levels. This methodology improvement decreases the actuarial balance by 0.10 percent of taxable payroll. As mentioned earlier, the net effect of this change in method and the change in the assumed ultimate total fertility rate is very small.
The second significant update is to the model used to project the civilian labor force by age, sex, marital status, and presence of children. As mentioned earlier under the “Economic Data and Assumptions” section, the model was updated to incorporate data from the latest complete economic cycle, thereby putting more weight on the recent relationships among the various factors affecting labor force participation. Overall, improvements to the labor force model decrease the actuarial balance by 0.04 percent of taxable payroll.
The third significant set of changes is related to the sample used for the long-range model for projecting average benefit levels of retired-worker and disabled-worker beneficiaries who become newly entitled for benefits. This model uses a large sample of 10 percent of all newly entitled retired-worker beneficiaries in a recent year. The sample used in last year’s report was for worker beneficiaries newly entitled in 2016, while this year’s report uses the results from worker beneficiaries newly entitled in 2017. This update to the sample data decreases the actuarial balance by 0.02 percent of payroll. Data updates and updates to the methodology for projecting average benefit levels combine to decrease the actuarial balance by an additional 0.07 percent of payroll. A primary reason for this decrease in actuarial balance is a method change to improve the assignment of projected earnings by age of initial entitlement, better reflecting the change between historical and projected retired worker distributions by entitlement age.
Fourth, recent data and estimates provided by the Office of Tax Analysis (OTA) at the Department of the Treasury indicate lower near-term and ultimate levels of revenue from taxation of OASDI benefits than projected in last year’s report. The decrease in projected ratios of income tax on benefits to benefit amounts decreases in the actuarial balance by 0.04 percent of taxable payroll.
Fifth, methodology changes to improve projections of retroactive benefits and effective interest rates combine to decrease the actuarial balance by 0.03 percent of payroll. The projections for retroactive benefits for retired workers now better capture the different rules for workers who become newly entitled prior to normal retirement age (NRA) versus those who become entitled at or after normal retirement age. In addition, the projections for retroactive benefits for all categories of beneficiaries were improved to incorporate additional years of historical data for determining retroactive amounts. Projections of effective interest rates were improved to better reflect the expected detailed operations of the separate trust funds in years after the short-range period and prior to trust fund depletion, and to more appropriately limit the rates used in years in which one trust fund becomes depleted and the other does not.
The sixth significant change is updating two sets of benefit adjustment factors based on new programmatic data: the post-entitlement adjustment factors and the Windfall Elimination Provision (WEP) reduction factors. Post-entitlement factors are used to account for changes in benefit levels, primarily due to differential mortality by benefit level and earnings after benefit entitlement. WEP reduction factors are used to adjust benefits for individuals who receive a pension based on specified categories of employment not covered by OASDI. These data updates combine to decrease the actuarial balance by 0.01 percent of payroll.
In addition to these six significant methodological changes and programmatic data updates, changes in starting levels and projected levels of OASI and DI beneficiaries and benefit amounts over the first 10 years of the projection period, updating other programmatic data, other small methodological improvements, and interactions among the various method changes and updates to programmatic experience combine to decrease the long-range actuarial balance by 0.03 percent of taxable payroll.
Figure IV.B4 compares the annual balances for this report and the prior year’s report for the combined OASDI program over the long-range (75-year) projection period. The figure illustrates the annual effects of the changes described earlier in this section.
The annual balances in this year’s report are lower (more negative) in years from 2021 through 2090, and then higher for 2091 through 2095. The differences in annual balances fluctuate somewhat in the early 2020s during the recovery from the recession, are small in the late 2020s, generally grow until the late 2050s, and then decline thereafter and change sign in 2090. For this year’s report, the annual balance improves from 2021 to 2022, due to the assumed path of recovery from the COVID-19 pandemic and the ensuing recession, which were not reflected in last year’s report. The trends later in the projection are mainly due to the changes made this year to fertility methodology and assumptions. In particular, the new cohort-based projection method extends the transition from current low birth rates to the higher ultimate total fertility rate of 2.00 children per woman, such that projected total fertility rates are lower in this report through calendar year 2036.
For 2095, the projected annual deficit is 4.34 percent of taxable payroll in this report, compared to 4.55 percent in last year's report. For the full 75-year projection period, the annual balances average 0.23 percentage point lower in this year’s report.

1
See appendix F.

2
Adjustments include adding deemed wage credits based on military service for 1983-2001 and reflecting the lower effective tax rates (as compared to the combined employee-employer rate) that apply to multiple-employer “excess wages.” Lower rates also applied to net earnings from self-employment before 1984 and to income from tips before 1988.

3
Amounts for 2020 and 2021 are adjusted to include in 2021 operations those benefit payments regularly scheduled in the law to be paid on January 3, 2021, which were actually paid on December 31, 2020 as required by the statutory provision for early benefit payments when the normal delivery date is on a weekend or holiday. Such shifts in payments across calendar years have occurred in the past and will occur periodically in the future whenever January 3rd falls on a Sunday. In order to provide a consistent perspective on trust fund operations over time, all trust fund operations in each year reflect the 12 months of benefits that are regularly scheduled for payment in that year.

4
For presentation in this report, if the scheduled January 3, 2021 payment, actually paid on December 31, 2020, were a) counted as reducing trust fund reserves at the end of 2020 and b) excluded from the calculation of 2021 cost, then the OASI trust fund ratio shown for 2021 would be 283 percent.

5
A program is solvent over any period for which the trust fund maintains a positive level of asset reserves. In contrast, the actuarial balance for a period includes the cost of having a target fund equal to 100 percent of the following year’s cost at the end of the period. Therefore, if a program ends the period with reserves that are positive but not sufficient to cover the following year’s costs, it will be solvent at the end of the period and yet still have a small negative actuarial balance for that period.

6
The 3.36 percentage point increase in the payroll tax rate required to achieve 75-year solvency differs somewhat from the 3.54 percent actuarial deficit. This is primarily because the rate increase required to achieve 75-year solvency reflects a zero trust fund reserve at the end of the period, whereas the 3.54 percent actuarial deficit incorporates an ending trust fund reserve equal to one year’s cost. While such an increase in the payroll tax rate would cause some behavioral changes in earnings and ensuing changes in benefit levels, such changes are not included in these calculations because they are assumed to have roughly offsetting effects on OASDI actuarial status over the 75-year long-range period as a whole.

7
The calculation of the payroll tax rate increase required to eliminate the actuarial deficit also does not include the effects of behavioral changes, because they are assumed to have roughly offsetting effects.

8
The present value of taxable payroll for 2021-95 is $591.5 trillion. The present value of GDP for 2021-95 is $1,698.0 trillion. In last year’s report, the present value of taxable payroll for 2020-94 was $554.6 trillion and the present value of GDP was $1,614.2 trillion.


Table of Contents Previous Next Tables Figures Index
SSA Home | Privacy Policy | Website Policies & Other Important Information | Site Map | Actuarial Publications August 31, 2021