The Windfall Elimination Provision (WEP) in Social Security: Comparing Current Law with Proposed Proportional Formulas

Social Security is a work-based federal insurance program that provides income support to workers and their eligible family members in the event of a worker’s retirement, disability, or death. About 6% of workers in paid employment or self-employment in 2019 were not covered by Social Security. A quarter of state and local government employees and most permanent civilian federal employees hired before January 1, 1984, were not covered, and these groups constituted the majority of noncovered workers.

For workers whose entire careers are covered by Social Security, the Social Security benefit formula is weighted to replace a greater share of career-average earnings (referred to as the replacement rate) for low-paid workers than for high-paid workers. However, providing an appropriate replacement rate for beneficiaries whose careers are split between covered and noncovered employment (referred to hereinafter as split-career beneficiaries) is challenging because years of noncovered earnings are marked as zeros in Social Security earnings records, so split-career beneficiaries appear to have low career-average earnings. Therefore, if there were no adjustment for noncovered earnings, split-career beneficiaries would receive a higher replacement rate than beneficiaries with the same earnings who spent their entire careers in covered employment. The windfall elimination provision (WEP) is a modified benefit formula that reduces certain retired or disabled workers’ Social Security benefits if they also have earnings not covered by Social Security and are entitled to pension benefits based on those noncovered earnings. The WEP aims to provide split-career beneficiaries with approximately the same replacement rate as similar workers whose entire careers were covered by Social Security.

Some have argued, however, that the current-law WEP formula generally fails to accurately adjust affected workers’ benefits. They say it overadjusts some affected workers’ benefits (i.e., it reduces them by too much), giving them a lower replacement rate than similar workers whose entire careers were covered by Social Security. In contrast, they argue it underadjusts some other affected workers’ benefits, giving them a higher replacement rate than similar workers whose entire careers were covered. Estimates in 2018 showed the current-law WEP overadjusted 69% of affected beneficiaries’ benefits and underadjusted for the remaining 31%.

Legislative proposals have been introduced to substitute the WEP with a proportional formula that would calculate Social Security benefits based on earnings from both covered and noncovered employment. The proportional formula’s supporters have argued it is a more accurate method to treat noncovered employment, because it would provide the same replacement rate for split-career beneficiaries and beneficiaries whose entire careers are covered by Social Security.

Compared with current law, a proportional formula would increase Social Security benefits for beneficiaries whose current-law WEP benefits are overadjusted and decrease benefits for those whose benefits are underadjusted. It would also decrease benefits for many beneficiaries with earnings from noncovered employment who are exempt from the current WEP reduction because they (1) have 30 or more years of substantial covered earnings, or (2) do not receive a pension based on noncovered earnings.

Proposals to establish a proportional formula have been discussed since the 1980s. However, applying the proportional formula requires a complete record of earnings from covered and noncovered employment, which were not readily available at that time. To obtain the complete earnings record, the Social Security Administration (SSA) would have needed a massive new operation system requiring extensive data reporting, maintenance, and correction processes, which could not have been accomplished quickly with limited costs. Therefore, the current-law WEP was enacted in 1983 as an approximate approach to adjust Social Security benefits for certain beneficiaries who had earnings in jobs not covered by Social Security. Today, SSA has 35 years of data on earnings from both covered and noncovered employment, implying that the proportional formula is now an option for Congress to consider. In 2019 (the 116th Congress), H.R. 3934 and H.R. 4540 would replace the current-law WEP approach with a proportional formula for certain individuals who would become eligible for Social Security benefits in 2022 or later.

The Windfall Elimination Provision (WEP) in Social Security: Comparing Current Law with Proposed Proportional Formulas

January 24, 2020 (R46194)
Jump to Main Text of Report

Contents

Summary

Social Security is a work-based federal insurance program that provides income support to workers and their eligible family members in the event of a worker's retirement, disability, or death. About 6% of workers in paid employment or self-employment in 2019 were not covered by Social Security. A quarter of state and local government employees and most permanent civilian federal employees hired before January 1, 1984, were not covered, and these groups constituted the majority of noncovered workers.

For workers whose entire careers are covered by Social Security, the Social Security benefit formula is weighted to replace a greater share of career-average earnings (referred to as the replacement rate) for low-paid workers than for high-paid workers. However, providing an appropriate replacement rate for beneficiaries whose careers are split between covered and noncovered employment (referred to hereinafter as split-career beneficiaries) is challenging because years of noncovered earnings are marked as zeros in Social Security earnings records, so split-career beneficiaries appear to have low career-average earnings. Therefore, if there were no adjustment for noncovered earnings, split-career beneficiaries would receive a higher replacement rate than beneficiaries with the same earnings who spent their entire careers in covered employment. The windfall elimination provision (WEP) is a modified benefit formula that reduces certain retired or disabled workers' Social Security benefits if they also have earnings not covered by Social Security and are entitled to pension benefits based on those noncovered earnings. The WEP aims to provide split-career beneficiaries with approximately the same replacement rate as similar workers whose entire careers were covered by Social Security.

Some have argued, however, that the current-law WEP formula generally fails to accurately adjust affected workers' benefits. They say it overadjusts some affected workers' benefits (i.e., it reduces them by too much), giving them a lower replacement rate than similar workers whose entire careers were covered by Social Security. In contrast, they argue it underadjusts some other affected workers' benefits, giving them a higher replacement rate than similar workers whose entire careers were covered. Estimates in 2018 showed the current-law WEP overadjusted 69% of affected beneficiaries' benefits and underadjusted for the remaining 31%.

Legislative proposals have been introduced to substitute the WEP with a proportional formula that would calculate Social Security benefits based on earnings from both covered and noncovered employment. The proportional formula's supporters have argued it is a more accurate method to treat noncovered employment, because it would provide the same replacement rate for split-career beneficiaries and beneficiaries whose entire careers are covered by Social Security.

Compared with current law, a proportional formula would increase Social Security benefits for beneficiaries whose current-law WEP benefits are overadjusted and decrease benefits for those whose benefits are underadjusted. It would also decrease benefits for many beneficiaries with earnings from noncovered employment who are exempt from the current WEP reduction because they (1) have 30 or more years of substantial covered earnings, or (2) do not receive a pension based on noncovered earnings.

Proposals to establish a proportional formula have been discussed since the 1980s. However, applying the proportional formula requires a complete record of earnings from covered and noncovered employment, which were not readily available at that time. To obtain the complete earnings record, the Social Security Administration (SSA) would have needed a massive new operation system requiring extensive data reporting, maintenance, and correction processes, which could not have been accomplished quickly with limited costs. Therefore, the current-law WEP was enacted in 1983 as an approximate approach to adjust Social Security benefits for certain beneficiaries who had earnings in jobs not covered by Social Security. Today, SSA has 35 years of data on earnings from both covered and noncovered employment, implying that the proportional formula is now an option for Congress to consider. In 2019 (the 116th Congress), H.R. 3934 and H.R. 4540 would replace the current-law WEP approach with a proportional formula for certain individuals who would become eligible for Social Security benefits in 2022 or later.


Introduction

Social Security provides insured workers and their eligible family members with a measure of protection against the loss of income due to the worker's retirement, disability, or death.1 The amount of the monthly benefit payable to workers and their family members is based on the worker's career-average earnings from jobs covered by Social Security (i.e., jobs in which the worker's earnings were subject to the Social Security payroll tax). Although participation in Social Security is compulsory for most workers, about 6% of all workers in paid employment or self-employment are not covered by Social Security.2 Most noncovered workers are state and local government employees who are covered by alternative staff-retirement systems or permanent civilian federal employees hired before January 1, 1984, most of whom are covered by the Civil Service Retirement System (CSRS) or other alternative retirement plans.3

Social Security benefits are designed to replace a certain percentage of a worker's career-average earnings (referred to as the replacement rate) for those who remain in covered employment throughout their careers. The benefit formula is weighted to replace a greater share of career-average earnings (i.e., provide a higher replacement rate) for low-paid workers than for high-paid workers. However, providing an appropriate replacement rate for beneficiaries whose careers are split between covered and noncovered employment (referred to hereinafter as split-career beneficiaries) is challenging because years of noncovered earnings are marked as zeros in Social Security earnings records, so split-career beneficiaries appear to have low career-average earnings. Therefore, without adjusting for noncovered earnings, split-career beneficiaries would receive a higher replacement rate than beneficiaries with the same earnings who spent their entire careers in Social Security-covered employment.

The windfall elimination provision (WEP) is a modified benefit formula that reduces Social Security benefits for certain retired or disabled workers who have earnings not covered by Social Security and are entitled to pension benefits based on those noncovered earnings (including certain foreign pensions).4 Its purpose is to remove an unintended advantage or windfall that these workers would otherwise receive as a result of the interaction between the regular Social Security benefit formula and the workers' relatively short careers in Social Security-covered employment. In December 2018, nearly 1.9 million people (or about 3% of all Social Security beneficiaries) were affected by the WEP.

Some argue that the current-law WEP formula generally fails to provide the correct benefit adjustment (reduction) to affected beneficiaries. It overadjusts the benefit for some affected workers by producing a relatively large benefit reduction that gives them a lower replacement rate than similar workers whose entire careers were covered by Social Security; in contrast, it underadjusts the benefit for some other affected beneficiaries by producing a relatively small benefit reduction, giving them a higher replacement rate than similar workers whose entire careers were covered by Social Security. Legislative proposals have been introduced to substitute the current WEP with a proportional formula that would provide the same replacement rate to split-career beneficiaries and beneficiaries whose entire careers are covered.

This report explains how the proportional formula would work and how it differs from the current-law WEP formula. It also discusses how Social Security benefits would change under the proportional formula for workers with different levels of earnings, years of noncovered earnings, and timing of those noncovered earnings (i.e., early career, midcareer, or late career). Lastly, this report concludes with historical and recent legislative proposals that are based on the proportional formula.

The Current-Law WEP

How Does the Current-Law WEP Work?

Among other requirements, a worker generally needs 40 earnings credits (10 years of Social Security-covered employment) to be eligible for a Social Security retired-worker benefit.5 The Social Security regular benefit formula applies three replacement factors—90%, 32%, and 15%—to three different brackets of a worker's average indexed monthly earnings (AIME), which is the monthly average of the 35 highest years of indexed covered earnings.6 The result is the primary insurance amount (PIA), which is the worker's basic benefit before any adjustments are made for factors such as cost-of-living adjustments (COLAs), early retirement, delayed retirement, or noncovered earnings. For workers who become eligible for benefits in 2020, the PIA is based on the formula in Table 1. The dollar amounts in the table, known as bend points, are adjusted annually for average earnings growth.

Table 1. Social Security Benefit Formula and the Current WEP Formula for Workers Who Attain Age 62, Become Disabled, or Die in 2020

(Regular formula factors versus the WEP formula factors)

Regular Formula

Current WEP Formula

Replacement Factors and Bend Points

Example: Regular PIA for AIME of $1,500

Replacement Factors and Bend Points

Example: WEP PIA for AIME of $1,500

90% of first $960 AIME

$864.00

40% of first $960 AIMEa

$384.00

32% of AIME over $960 and through $5,785

172.80

32% of AIME over $960 and through $5,785

172.80

15% of AIME over $5,785

0.00

15% of AIME over $5,785

0.00

Total Benefitsb

$1,036.80

Total Benefits

$556.80

Source: Congressional Research Service (CRS), based on Social Security Administration (SSA), Benefit Formula Bend Points and Windfall Elimination Provision.

a. The first replacement factor increases from 40% to 90% when years of coverage (YOCs) increase from 20 to 30 by an increment of 5% for each additional year of coverage. YOCs are defined below.

b. The PIA (total benefit) is rounded down to the nearest 10 cents.

Under current law, the WEP reduction is based on years of coverage (YOCs)—the larger the number of YOCs, the lower the WEP reduction. For people with 20 or fewer YOCs who become eligible for benefits in 2020, the WEP reduces the first replacement factor from 90% to 40% (referred to as the WEP replacement factor in this report), resulting in a maximum benefit reduction of $480 (90% of $960 minus 40% of $960). A worker with an AIME of $1,500 who becomes eligible for Social Security benefits in 2020 would receive an unadjusted monthly benefit of $1,036.80 if all earnings are covered by Social Security, compared to a WEP-reduced monthly benefit of $556.80 if he or she has 20 or fewer YOCs (see Table 1). For each YOC in excess of 20, the WEP replacement factor increases by 5%. For example, the WEP factor is 45% for those with 21 YOCs and 50% for those with 22 YOCs. The WEP factor reaches 90% for those with 30 or more YOCs, and at that point it is phased out (see Figure 1).

The amount of substantial covered earnings needed for a YOC is $25,575 in 2020; the amount is adjusted annually by average wage growth.7 Workers with annual covered earnings below the level of substantial earnings do not receive a YOC. For example, a worker who earns $5,640 in 2020 (covered earnings) will receive four earnings credits for the purpose of Social Security eligibility, but will not qualify for a YOC for the WEP purpose. In December 2018, of the nearly 1.9 million beneficiaries affected by the WEP, nearly 1.6 million (84%) had 20 YOCs or fewer, and the remaining 0.3 million (16%) had 21-29 YOCs (see Figure 1).

Two groups of beneficiaries with noncovered employment are exempt from the WEP: (1) those with 30 or more YOCs; and (2) those not receiving a pension based on those noncovered earnings. SSA's Office of the Chief Actuary estimated that roughly 18 million Social Security worker beneficiaries with some noncovered earnings were exempt from the current WEP in 2018.8 Among them, about 9.4 million (52%) had 30 or more YOCs. Additionally, a guarantee provision in the WEP ensures that the WEP reduction cannot exceed one-half of the pension based on the worker's noncovered employment.

Figure 1. The WEP Replacement Factor Under Current Law and Number of Beneficiaries Affected by the WEP in December 2018, by YOCs

Source: Social Security Administration (SSA), Windfall Elimination Provision, at https://www.ssa.gov/pubs/EN-05-10045.pdf; and unpublished data received by CRS from SSA, Office of Research, Evaluation, and Statistics (ORES).
Notes: A worker needs to earn a substantial amount in covered employment to qualify for a YOC. The amount of substantial covered earnings needed for a YOC is $25,575 in 2020. A worker with many years of earnings below the YOC threshold would have zero YOC.

Benefit Adjustments Under the Current-Law WEP

The regular Social Security benefit formula is progressive, replacing a greater share of career-average earnings for low-paid workers than for high-paid workers. For example, Table 2 displays five types of scaled workers with hypothetical lifetime earnings from low to high, whose earnings patterns are based on actual Social Security-insured workers' career earnings.9 The replacement rate—the percentage of AIME replaced by the PIA—ranges from 83.3% for a very low-earning worker whose entire career is covered to 60.5% for a low-earning worker, 44.8% for a medium-earning working, 37.2% for a high-earning worker, and 29.4% for a worker who earns the taxable maximum every year.10

If a person has earnings not covered by Social Security, those noncovered earnings are shown as zeros in their Social Security earnings records, thus resulting in relatively lower career-average earnings. The regular formula cannot distinguish between workers who have low career-average earnings because they worked for many years at low earnings in covered employment and workers who appear to have low career-average earnings because they worked for many years in jobs not covered by Social Security. Therefore, without a PIA reduction for noncovered earnings, a worker who split his or her career between covered and noncovered employment might receive a higher replacement rate than a worker with the same level of earnings who spent an entire career in covered employment. For example, a low-scaled worker is estimated to have annual career-average earnings of $22,588. If all career earnings were covered, the worker would receive a 60.5% replacement rate in Social Security benefits. However, if the second half of the low-scaled worker's career was in noncovered employment, the worker would receive a replacement rate of 90.0% based on the regular benefit formula before adjusting for noncovered earnings (see Table 2). The WEP PIA addresses this problem by reducing the replacement rate for certain workers who have noncovered earnings. For example, the replacement rate would be adjusted from 90.0% to 40.0% for a low-scaled worker if the second half of his career was not covered by Social Security.

Table 2. Replacement Rates (PIA/Covered AIME) for Hypothetical Scaled Workers

(Workers are eligible for benefits in 2020)

 

Scaled Workers

 

Very Low

Low

Medium

High

Taxable Maximum

Annual career-average earnings (covered and noncovered)

$12,549

$22,588

$50,196

$80,313

$123,232

Replacement Rates for

 

 

 

 

 

All earnings covered by Social Security

83.3%

60.5%

44.8%

37.2%

29.4%

Employment between ages 21 and 40 covered by Social Security, between ages 41 and 61 not covered

 

 

 

 

 

Without WEP

90.0%

90.0%

60.2%

49.6%

41.4%

Current WEP PIA

40.0%

40.0%

35.9%

34.4%

33.2%

Source: CRS. Scaled workers and their career average earnings are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: Hypothetical workers are assumed to work from age 21 to age 61. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. Career-average earnings are defined as the average of the highest 35 years of covered earnings, indexed for growth in average wages. Replacement rates are calculated using Regular or WEP PIA divided by AIME based on earnings covered by Social Security.

The WEP's original intent was to ensure that Social Security beneficiaries with some earnings from noncovered employment received the same replacement rate as workers who spent their entire careers in covered employment.11 However, the current-law WEP formula can only approximately achieve that goal.

The current-law WEP formula overadjusts benefits for certain affected beneficiaries by producing a relatively large benefit reduction, resulting in a lower replacement rate than a similar worker whose entire career was covered by Social Security would receive. For example, a very low-scaled worker who spent the second half of his or her career in noncovered employment would receive a replacement rate of 40.0% using the WEP formula, which is substantially lower than the replacement rate a very low-scaled worker whose entire career was covered by Social Security (83.3%) would receive. The magnitude of such benefit overadjustment is smaller for affected beneficiaries with relatively higher lifetime earnings. For example, if the second half of a high-scaled worker's career was not covered by Social Security, the worker would receive a WEP benefit replacing 34.4% of covered AIME, which is slightly lower than the replacement rate for high-scaled workers whose entire careers are covered by Social Security (37.2%).

In addition, the current-law WEP formula underadjusts Social Security benefits for some other beneficiaries by producing a relatively small benefit reduction, resulting in a higher replacement rate than a similar worker whose entire career is covered would receive. Such underadjustment usually applies to workers with significantly high lifetime earnings and some earnings not covered by Social Security. For example, a taxable-maximum worker who earned the taxable-maximum amount each year of work history is estimated to have career-average earnings of $123,232. If the second half of the taxable-maximum worker's career was not covered by Social Security, the worker would receive a 33.2% replacement rate in Social Security benefits under the WEP, compared to 29.4% if the entire career had been covered by Social Security (see Table 2).

Scaled Workers

The Social Security Administration's (SSA's) Office of the Chief Actuary (OCACT) developed scaled worker hypothetical earnings patterns for four different career-average earnings levels: very low, low, medium, and high. These patterns express hypothetical earnings at each age as a percentage of the SSA's national average wage index (AWI). The scaled factors used to determine the scaled earnings patterns are based on the average work and earnings of actual insured workers over their careers. At each age, the scaled factor reflects both the average earnings level of those working at that age and the percentage of insured workers working at that age.12

Based on the assumptions for scaled workers, those born in 1958 would have hypothetical lifetime earnings as shown in Figure 2. A typical scaled worker's earnings start off relatively low in the early years and then increase gradually over time. Earnings generally peak when a scaled worker enters his or her 50s and then decline as the worker approaches retirement age. The figure also shows a taxable-maximum worker's earnings. This worker would earn the taxable-maximum amount in each year of work history.

Figure 2. Annual Earnings for Scaled Workers by Age

(Worker born in 1958)

Source: CRS analysis based on Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: Workers born in 1958 turned age 62 in 2020, which is the year when they first become eligible for Social Security retirement benefits. From 2013 to 2018, 18.3% of actual retiring workers had AIME closest to the very low-scaled earnings level, 22.5% were closest to the low-scaled level, 29.9% were closest to the medium-scaled level, 20.4% were closest to the high-scaled level, and 8.8% were closest to the taxable maximum.

The Proportional Formula for the WEP

How Would the Proportional Formula Work?

The proportional formula for the WEP would apply the regular Social Security benefit formula to all past earnings up to the taxable maximum from both covered and noncovered employment. The resulting benefit would then be multiplied by the ratio of career-average earnings (AIME) from covered employment only to career-average earnings (AIME) from both covered and noncovered employment. By concept, the PIA under the proportional formula (i.e., proportional PIA) would be as follows:

Proportional PIA=PIA for all Earnings×AIME for Covered EarningsAIME for all Earnings

In other words, a Social Security benefit would be calculated based on a worker's combined covered and noncovered earnings, but only the portion based on covered earnings would be payable as a Social Security benefit.

Benefit Adjustments Under the Proportional Formula

Under the proportional formula, Social Security beneficiaries with some earnings from noncovered employment would receive the same replacement rate (ratio of PIA to AIME) for covered earnings as similarly situated workers who spent their entire careers in covered employment, regardless of earnings levels, years of covered earnings, or the timing of those covered earnings. Figure 3 illustrates this, showing that a medium-scaled worker would receive a 44.8% replacement rate under the proportional formula whether the worker's entire career or only half of the worker's career was covered by Social Security. This 44.8% replacement rate for the split-career worker would be lower than the windfall replacement rate under the regular PIA without any adjustment for noncovered earnings (60.2%), but higher than the rate under the current WEP PIA (35.9%), which overadjusts the benefit reduction for noncovered earnings.

Figure 3. Replacement Rates (PIA/Covered AIME) Under Alternative Formulas for Medium-Scaled Workers

(Workers are eligible for benefits in 2020)

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: Hypothetical workers are assumed to work from age 21 to age 61. The worker who split his or her career between covered and noncovered employment is assumed to work in jobs covered by Social Security between the ages of 21 and 40, and in jobs not covered by Social Security between the ages of 41 and 61. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. Replacement rates are calculated using Regular PIA, WEP PIA, or Proportional PIA divided by AIME based on earnings covered by Social Security.

The proportional formula would provide a higher benefit than the WEP for workers whose Social Security benefits are currently overadjusted, such as the very low-, low-, medium-, and high-scaled workers shown in Table 3. Because scaled workers with relatively lower lifetime earnings receive a larger overadjustment under the current WEP, those workers would receive a larger monthly benefit increase under the proportional formula. For example, the monthly benefit increase under the proportional formula relative to the current WEP would be $213.90 for very low-scaled workers if their careers' second halves were not covered by Social Security, compared to $182.20 for low-scaled workers, $176.50 for medium-scaled workers, and $87.10 for high-scaled workers. In contrast, workers whose Social Security benefits are underadjusted by the current WEP, such as taxable-maximum workers, would receive a lower benefit under the proportional formula.

Table 3. Replacement Rates (PIA/Covered AIME) and Benefit Differences Between Alternative Formulas for Hypothetical Scaled Workers

(Workers are eligible for benefits in 2020)

 

Scaled Workers

 

Very Low

Low

Medium

High

Taxable Maximum

All earnings covered by Social Security

83.3%

60.5%

44.8%

37.2%

29.4%

Earnings between ages 21 and 40 covered by Social Security, between ages 41 and 61 not covered

 

 

 

 

 

Current WEP PIA

40.0%

40.0%

35.9%

34.4%

33.2%

Proportional PIA

83.3%

60.5%

44.8%

37.2%

29.4%

Monthly Benefit Differences
(Proportional PIA-WEP PIA)

$213.9

$182.2

$176.5

$87.1

-$218.9

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: Hypothetical workers are assumed to work from age 21 to age 61. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. Replacement rates are calculated using Regular PIA, WEP PIA, or Proportional PIA divided by AIME based on earnings covered by Social Security.

Comparing the Proportional Formula with Current Law

The proportional formula discussed above would differ from the current-law WEP formula in terms of monthly benefit amounts, improper payments, and notification to beneficiaries.

Differences in Monthly Benefits

Given the current-law WEP formula's design, the proportional formula would increase Social Security benefits for some beneficiaries with noncovered employment and decrease benefits for others. Beneficiaries who would receive a lower benefit under the proportional formula than under current law include

  • beneficiaries with noncovered earnings who are exempt from the current-law WEP, such as those with 30 or more YOCs or those not receiving a noncovered pension; and
  • beneficiaries whose benefits are underadjusted using the current WEP PIA, such as those who have relatively high lifetime earnings, are close to 30 YOCs, or are affected by the current-law guarantee provision.

If the proportional formula had applied to current beneficiaries in 2018, SSA's Office of the Chief Actuary (OCACT) estimates that about 1.1 million beneficiaries affected by the current WEP (or 69%) would have received a higher benefit and about 0.5 million beneficiaries affected by the current WEP (or 31%) would have received a lower benefit. In addition, 13.5 million beneficiaries with some noncovered earnings who were exempted from the current WEP in 2018 would have received a lower benefit under the proportional formula.13

Exemptions and the Guarantee Provision Under Current Law

Beneficiaries who are eligible for either of the two exemptions to the current-law WEP would receive a lower benefit under the proportional formula. Beneficiaries with 30 or more YOCs are exempted from the current WEP, but under the proportional formula, workers with 30 or more YOCs and very few years of noncovered employment (even less than a year) would probably receive proportional reductions in their Social Security benefits. For example, a medium scaled-worker who earned 30 YOCs in his earlier career would not be affected by the current WEP even if he took a noncovered position afterward and was entitled to a noncovered pension (see case [1] in Table 4). In this case, the worker would receive an unreduced Social Security benefit of $1,707.30, which would be higher than the proportional PIA ($1,612.00) based on earnings from noncovered employment. SSA's OCACT estimates that, in 2018, roughly 9.4 million Social Security retired-worker and disabled-worker beneficiaries with some noncovered earnings were exempt from the current WEP because they had 30 or more YOCs. Because those beneficiaries have relatively few years of noncovered earnings, their benefit reductions under the proportional formula would be relatively small.14

The other exemption applies to beneficiaries with noncovered earnings who do not receive a pension based on those noncovered earnings. Those beneficiaries could receive a lower benefit under the proportional formula because their earnings from noncovered employment could reduce the proportion of overall career-average earnings from covered jobs.15 For example, under current law, a medium-scaled worker who worked in a noncovered position from age 55 to age 61 but received no noncovered pension benefits would be exempt from the WEP and receive a Social Security benefit equal to $1,551.10 (see case [2] in Table 4). This amount would be higher than the benefit computed by the proportional formula ($1,432.80) because those seven years of noncovered employment would proportionally reduce the Social Security benefit. Estimates from OCACT find that about 8.6 million Social Security retired-worker and disabled-worker beneficiaries with some noncovered earnings and less than 30 YOCs were exempt from the current WEP in 2018 because they had no pension based on those noncovered earnings.16

In addition, the guarantee provision under current law limits benefit reductions by ensuring that the WEP reduction cannot exceed one-half of the noncovered pension benefit. This provision typically leads to small benefit reductions for beneficiaries who receive small pension benefits based on relatively short careers in noncovered employment. The proportional formula would not limit reductions in this way, so those workers' benefits would be lower under the proportional formula than under the WEP. For example, a low-scaled worker who worked in a noncovered position from age 52 to age 61 and received a monthly benefit from a noncovered pension equal to $100 would receive a WEP reduction of no more than $50 under current law (see case [3] in Table 4). Therefore, this worker would receive $896.10 under the current WEP, but $770.90 under the proportional formula with no guarantee provision.

Table 4. Monthly Benefits Based on the Current WEP and the Proportional Formula for Certain Medium- or Low-Scaled Workers

(Worker eligible for benefits in 2020)

Employment History

Monthly Noncovered Pension Benefits

Current-Law WEP

Proportional Formula

Treatment Under WEP

(1) A medium-scaled worker: Aged 21-52 covered; aged 53-61 noncovered

$500

$1,707.30

$1,612.00

Exempt for 30 YOCs

(2) A medium-scaled worker: Aged 21-30 not working; aged 31-54 covered; aged 55-61 noncovered

None

$1,551.10

$1,432.80

Exempt for no pension

(3) A low-scaled worker: Aged 21-30 not working; aged 31-51 covered; aged 52-61 noncovered

$100

$896.10

$770.90

Guarantee provision

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. For a medium-scaled worker, earnings at ages 21 and 22 would be lower than the substantial earnings requirement for WEP purposes; for a low-scaled worker, earnings at ages 21 to 37 and ages 56 to 61 would be lower than the substantial earnings. Therefore, the number of years of substantial covered earnings (YOCs) is less than the number of years in covered employment for medium-scaled workers and low-scaled workers. For example, a medium-scaled worker who worked from age 21 to age 52 in covered employment and the remaining years in noncovered employment would have 30 as opposed to 32 YOCs.

Years of Coverage

In addition to the exemptions and the guarantee provision, whether a worker with noncovered earnings would receive a lower Social Security benefit under the proportional formula relative to current law also depends on YOCs based on substantial earnings. The number of YOCs determines the WEP replacement factor under current law (see Figure 1). In general, the larger the number of YOCs, the higher the WEP replacement factor. Workers who have employment not covered by Social Security also need to earn the substantial covered amount ($25,575 in 2020) to receive one YOC, which is much higher than the earnings required for Social Security eligibility ($5,640 in 2020). Because of the WEP's higher YOC earnings threshold, workers with relatively lower covered earnings who are affected by the WEP may be entitled to Social Security benefits based on earnings credits but not qualify for a YOC for WEP purposes.

Although YOCs are a critical factor for determining the PIA under the current-law WEP, they are not relevant for the proportional formula. To compare monthly benefits based on the two formulas by YOCs, Figure 4 shows a medium-scaled worker's monthly benefit amounts under the current WEP PIA and the proportional PIA. If the medium-scaled worker took a job covered by Social Security in the earlier part of her career and the number of YOCs was relatively small (less than 27 for a medium-scaled worker), the proportional formula would provide a higher benefit than the current WEP. However, if the number of YOCs were relatively large (more than 27 for a medium-scaled worker), the proportional formula would provide a lower benefit than the current–law WEP.

Two reasons may explain why the proportional formula would provide a lower benefit than the current-law WEP at the higher level of YOCs. First, current law exempts beneficiaries from the WEP if they have 30 or more YOCs, resulting in a higher benefit amount than under the proportional formula. Second, when YOCs are close to 30, the current-law WEP replacement factor is relatively large, such as 85% for 29 YOCs (see Figure 1), so the current-law WEP PIA underadjusts and produces a higher benefit than the proportional formula.

Figure 4. Monthly Benefits and Benefit Differences Based on Current WEP PIA and Proportional PIA for A Medium-Scaled Worker

(Worker eligible for benefits in 2020)

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: The medium-scaled worker is assumed to work from age 21 to age 61, with earlier years of employment covered by Social Security and later years of employment not covered. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. For a medium-scaled worker, his or her earnings at ages 21 and 22 would be lower than the substantial earnings requirement for a YOC under the WEP. Therefore, the number of years of substantial covered earnings is two years fewer than the number of years in covered employment. If, for example, this medium-scaled worker worked from age 21 to age 30 in covered employment and remaining years in noncovered employment, the worker would have eight years of substantial covered earnings.

The monthly benefit difference between the proportional formula and the current-law WEP formula also depends on earning levels. For example, the very low- and low-scaled workers in Figure 5 had fewer than 20 YOCs because their annual earnings were typically less than the substantial earnings required for a YOC. They would receive a current-law WEP PIA based on the lowest WEP replacement factor (40%).17 Therefore, the proportional PIA for these workers would generally be higher than the WEP PIA, because the 40% WEP replacement factor overreduces their Social Security benefits for noncovered earnings. The proportional PIA would also be higher than the WEP PIA for medium- and high-scaled workers with relatively fewer YOCs, such as Figure 5's medium-scaled workers with fewer than 29 years of covered earnings and high-scaled workers with YOCs between 11 and 22.18 However, as YOCs increase, the WEP replacement factor goes up, so the WEP PIA is higher than the proportional PIA for medium- and high-scaled workers with more YOCs. For workers with substantially high earnings, such as taxable maximum workers, the proportional PIA would generally be lower than the WEP PIA.

Figure 5. Monthly Benefit Differences Between Proportional PIA and Current WEP PIA for Scaled Workers

(Worker eligible for benefits in 2020)

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: The scaled worker is assumed to work from age 21 to age 61, with earlier years of employment covered by Social Security and later years of employment not covered. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement. In the figure, positive amounts reflect higher benefits under the proportional formula and negative amounts reflect higher benefits under the current WEP formula.

Timing of Noncovered Employment

The size of the monthly benefit difference between the proportional PIA and the WEP PIA also depends on the timing of covered and noncovered employment. Figure 6 compares three medium-scaled workers with 20 years of covered employment in early career, midcareer, and late career, respectively. Because early-career earnings are relatively lower than earnings in later years, a medium-scaled worker whose early career is covered by Social Security would tend to have a lower WEP PIA, a lower proportional PIA, and a lower monthly benefit difference between the two formulas than a medium-scaled worker with covered earnings at midcareer or late career. This example indicates that the WEP PIA and proportional PIA amounts depend on the timing of noncovered employment, as well as earning levels from both covered and noncovered employment.

Figure 6. Monthly Benefits Under Current WEP PIA and Proportional PIA for Medium-Scaled Workers, by Timing of Covered Employment

(Shaded periods indicate employment covered by Social Security)

Source: CRS. Scaled workers are determined using Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

Notes: The medium-scaled worker is assumed to work from age 21 to age 61, and become eligible for Social Security in 2020. Benefits are evaluated at age 62, and no actuarial reduction is taken for early retirement.

Administration and Improper Payments

The current-law WEP and the proportional formula differ not only in benefit calculation, but also in administration and associated costs.

SSA's ability to administer the current WEP depends in large part on the type of noncovered employment on which a beneficiary's pension is based. For most federal retirees and survivors, SSA relies primarily on noncovered pension data matched from the Office of Personnel Management (OPM).19 However, for state or local retirees and certain retirees with foreign pensions, SSA relies primarily on beneficiaries to self-report noncovered pension amounts.20 Based on the information matched and provided, SSA determines whether and to what extent to apply the WEP. Unreported state and local government pensions lead to improper payments. According to SSA, WEP has been a leading cause of computational errors related to overpayments.21 For FY2013 through FY2017, WEP accounted for 63% of reported computation overpayment errors, and average overpayments related to WEP totaled approximately $520 million annually.22

In contrast, the proportional formula is applied based on covered and noncovered earnings records, which are reported to SSA on Internal Revenue Service (IRS) Form W-2.23 Without other provisions, benefits based solely on the proportional formula would likely have fewer errors compared to benefits computed with the current-law WEP formula.

Notification to Beneficiaries

The annual Social Security statements that SSA makes available to all eligible workers provide benefit estimates based only on covered employment, with no estimates of the WEP adjustment because SSA is not provided with information on receipt of noncovered pensions until an individual self-reports this benefit when applying for Social Security. Because of this limitation, beneficiaries have argued that they were not given sufficient notice of how much their benefits would be reduced due to the WEP.24 To address this issue, the Social Security Protection Act of 2004 (P.L. 108-203) requires state and local government employers to disclose the WEP's effect to affected employees hired on or after January 1, 2005. SSA also responded to those communication issues by inserting a description of the WEP into the statement beginning in 2007.

However, communication challenges remain. The statement provides no estimates of the current WEP adjustment. The WEP adjustment is difficult to estimate without information on noncovered pensions, which is generally not available until the worker is entitled to such pension at a later date.

Compared to the current WEP, the estimate of noncovered earnings used in the proportional formula and the corresponding proportional PIA would be relatively easier to include in the statement. The proportional PIA estimate would have to be based on certain assumptions regarding future employment type, but it would not require noncovered pension information.

Legislative Proposals Based on the Proportional Formula

Proposals in the 1980s

1981

In 1981, proposals to address Social Security benefits for individuals receiving pensions from noncovered employment were discussed as part of broad reform efforts to address Social Security's financing issues, which were a major concern at the time.25

Some of the proposals called for worker PIA computations to use both covered and noncovered earnings, and for the PIA based on combined earnings to then be reduced by the ratio of noncovered earnings to combined earnings. This method is commonly referred to as the proportional formula, as discussed earlier in this report. This proposal was recommended by the National Commission on Social Security26 and included in Section 301 of H.R. 3207, the Social Security Amendments of 1981 as introduced in the 97th Congress.27

Other proposals called for a modified benefit formula that would change the first replacement factor in the regular benefit formula for workers with pensions based on noncovered work, which is similar to current law. For example, a May 1981 Reagan Administration proposal would have substituted the 90% replacement factor in the regular benefit formula with a 32% replacement factor for affected beneficiaries. The proposal would have guaranteed that the Social Security benefit reduction could not exceed one-half of the noncovered pension.

1983

In January 1983, the National Commission on Social Security Reform (NCSSR, better known as the Greenspan Commission) recommended eliminating the windfall portion of benefits for individuals who received a pension based on noncovered employment. The two methods discussed above were suggested: (1) the proportional formula based on covered and noncovered earnings, and (2) the modified benefit formula, substituting the 90% replacement factor with 32%.28

In the same year, SSA offered comments on the two methods. The agency indicated that the proportional formula would be the most conceptually appropriate, but would require SSA to maintain detailed records on workers' noncovered earnings in a manner comparable to the current covered earnings record operations, which would have required extensive data reporting, maintenance, and correction processes, and could likely not have been done with limited cost at that time. In contrast, SSA indicated the modified benefit formula based on the replacement factor would achieve the proportional formula's approximate results and be vastly easier to administer. SSA also recommended lowering the 90% replacement factor to 61% (the midpoint between the 90% factor and the 32% factor), as the 32% replacement factor would overadjust for the windfall.29

In March 1983, Congress incorporated the NCSSR's recommendations (with some modifications), along with additional provisions to resolve the remaining long-range deficit, into the Social Security Amendments of 1983 (P.L. 98-21). The conference agreed that the 90% replacement factor in the regular benefit formula would be substituted with a 40% replacement factor (phased in over five years), as in current law.30

Proposals from 2004 to Present

Since 2004, various bills have been introduced to replace the current WEP formula with the proportional formula based on both covered and noncovered earnings.31 Partly because all covered and noncovered earnings have been reported to SSA on Form W-2 since 1978, sufficient earnings records are now available to apply the proportional formula. Thus, a previous major area of concern for administering a proportional formula has been alleviated.

Legislative proposals based on the proportional formula usually address two essential questions: (1) whether the proportional formula would be applied to beneficiaries affected by the current WEP; and (2) how to treat beneficiaries who would receive a lower benefit under the proportional formula compared to current law. For the first question, proposals either apply the proportional formula to all current and future affected beneficiaries, or apply the proportional formula only to certain future beneficiaries and provide an additional monthly benefit (usually referred to as a rebate) to those affected by the current WEP. For the second question, some proposals include a no-benefit-cut provision such that the beneficiary would receive a benefit based on the higher of the current WEP formula and the proportional formula.

For example, S. 113 and H.R. 2797 in the 112th Congress would have applied the proportional formula to all beneficiaries (both current and future beneficiaries) after 1985 and provided a no-benefit-cut or hold harmless provision to beneficiaries who had worked in noncovered positions prior to one year after the bill's enactment. In a somewhat different approach, H.R. 3934 and H.R. 4540 in the 116th Congress would apply the proportional formula to beneficiaries becoming eligible after a certain date, such as December 31, 2021; provide a rebate to beneficiaries affected by the current-law WEP; and mandate a no-benefit-cut provision for some or all future beneficiaries.32 The above two bills introduced in the 116th Congress also include provisions to require SSA to show noncovered as well as covered earnings records on Social Security statements and to require studies on ways to facilitate data exchanges between SSA and state and local governments to improve current-law WEP administration.

Author Contact Information

Zhe Li, Analyst in Social Policy ([email address scrubbed], [phone number scrubbed])

Footnotes

1.

See CRS Report R42035, Social Security Primer, by Barry F. Huston.

2.

Social Security Administration (SSA), Fact Sheet on the Old-age, Survivors, and Disability Insurance Program, August 5, 2019, at https://www.ssa.gov/OACT/FACTS/fs2019_06.pdf.

3.

See CRS In Focus IF10243, Civilian Federal Retirement: Current Law, Recent Changes, and Reform Proposals, by Katelin P. Isaacs.

4.

For more information, see CRS Report 98-35, Social Security: The Windfall Elimination Provision (WEP), by Zhe Li. The WEP does not apply to workers whose only noncovered pension is based on earnings from noncovered domestic or foreign employment before 1957; or workers who receive foreign pension payments after 1994 that are based on a totalization agreement with the United States.

5.

A worker may earn up to four earnings credits per calendar year. In 2020, a worker earn one credit for each $1,410 of covered earnings, up to a maximum of four credits for covered earnings of $5,640 or more. Earnings credits are also called quarters of coverage.

6.

For detailed information on computing AIME, see CRS Report R43542, How Social Security Benefits Are Computed: In Brief, by Barry F. Huston. In short, the AIME computation process first updates past earnings to account for growth in overall economy-wide earnings by increasing each year of a worker's taxable earnings after 1950 by the growth in average earnings in the economy, as measured by the national average wage index, from the year of work until two years before eligibility for benefits, which for retired workers is at 62. For retired workers, the AIME equals the average of the 35 highest years of indexed earnings, divided by 12 (to change from an annual to a monthly measure). If the person worked fewer than 35 years in employment subject to Social Security payroll taxes, the computation includes some years of zero earnings.

7.

This amount is adjusted annually by the growth in average earnings in the economy, provided a cost-of-living adjustment (COLA) is payable. For more information, see Social Security Administration (SSA), Windfall Elimination Provision, at https://www.ssa.gov/pubs/EN-05-10045.pdf.

8.

Letter from Stephen C. Goss, Chief Actuary, SSA, to the Honorable Kevin Brady, U.S. House, July 24, 2019, https://www.ssa.gov/oact/solvency/KBrady_20190724.pdf, and letter from Stephen C. Goss, Chief Actuary, SSA, to the Honorable Richard Neal, U.S. House, September 30, 2019, https://www.ssa.gov/oact/solvency/RNeal_20190930.pdf. The projections are based on the intermediate assumptions of the 2019 Social Security trustees report (hereinafter "SSA OCACT Letters 2019").

9.

The definition of scaled workers is explained in the text box following this section.

10.

The taxable maximum, or taxable earnings base, is $137,700 in 2020, and increases annually with national average wage growth. This is the maximum earnings amount that is subject to Social Security taxes. For more information, see CRS Report RL32896, Social Security: Raising or Eliminating the Taxable Earnings Base, by Zhe Li.

11.

SSA, Appendix C of the 1983 Greenspan Commission on Social Security Reform, 1983, at https://www.ssa.gov/history/reports/gspan5.html.

12.

For more information about scaled workers, see Michael Clingman and Kyle Burkhalter, Scaled Factors for Hypothetical Earnings Examples under the 2019 Trustees Report Assumptions, SSA, Actuarial Note Number 2019.3, April 2019, at https://www.ssa.gov/OACT/NOTES/ran3/an2019-3.pdf.

13.

SSA OCACT Letters 2019.

14.

SSA OCACT Letters 2019.

15.

If the noncovered pension is only based on employee contributions, it is subject to the WEP only if it is the primary retirement plan. Therefore, beneficiaries with only a noncovered supplementary pension would be exempt from the current WEP, and their benefits under the proportional formula would be lower than under current law. For more information, see SSA, Program Operations Manual System, RS 00605.364 at https://secure.ssa.gov/poms.nsf/lnx/0300605364.

16.

SSA OCACT Letters 2019.

17.

A low-scaled worker born in 1957 who worked from age 21 to age 61 would have 18 years of earnings higher than the substantial earnings levels under the WEP, and a very low-scaled worker working during the same time would have no annual earnings higher than the substantial earnings levels.

18.

For a medium-scaled worker, his or her earnings at ages 21 and 22 would be lower than the substantial earnings requirement for a YOC under the WEP. Therefore, the number of years of substantial covered earnings is two years fewer than the number of years in covered employment. If, for example, this medium-scaled worker worked from age 21 to age 49 (29 years) in covered employment and remaining years in noncovered employment, the worker would have 27 years of substantial covered earnings.

19.

Social Security Administration, "Privacy Act of 1974; Matching Program," 84 Federal Register 2293, February 6, 2019, https://www.govinfo.gov/content/pkg/FR-2019-02-06/pdf/2019-01198.pdf.

20.

Social Security Administration, Full Justification of Estimates for Appropriations Committees, FY2017, February 9, 2016, p. 23, https://www.ssa.gov/budget/.

21.

Social Security Administration, SSA's FY2018 Agency Financial Report, p. 184, at https://www.ssa.gov/finance/2018/Full%20FY%202018%20AFR.pdf.

22.

SSA's OIG, Audit Report: Windfall Elimination Provision Exemptions, August 2019, at https://oig.ssa.gov/sites/default/files/audit/full/pdf/A-13-17-34132.pdf.

23.

Certain earnings from foreign countries may not be properly recorded by U.S. agencies.

24.

Social Security Advisory Board (SSAB), The Windfall Elimination Provision: It's Time to Correct the Math, October 1, 2015.

25.

See CRS Report RL30920, Social Security: Major Decisions in the House and Senate Since 1935, by Tamar B. Breslauer and William R. Morton.

26.

National Commission on Social Security, Social Security in America's Future: Final Report of the National Commission on Social Security, March 1981, Chapter 8: "Coverage of Social Security," pp. 191-192.

27.

The bill was introduced by Rep. J.J. Pickle (Chairman of the House Ways and Means Social Security Subcommittee) in April 1981.

28.

For more information, see National Commission on Social Security Reform, Report of the National Commission on Social Security Reform, January 1983, pp. 2-10.

29.

Unpublished SSA document, January 21, 1983, pp. 1-2.

30.

The House-passed version of the Social Security Amendments of 1983 (H.R. 1900) proposed to substitute the 90% replacement factor in the regular benefit formula with a 61% replacement factor, and the Senate version of the bill would have substituted the 90% replacement factor in the regular benefit formula with a 32% replacement factor (with a five-year phase-in). The conferees agreed to set the replacement factor at 40% as a compromise. The bill also required that workers with 30 or more years of Social Security coverage were exempt from the provision; that for workers with 26 to 29 years of coverage, a factor larger than 40% was applied (on a sliding scale); and that the reduction in the worker's Social Security benefit could not be more than one-half of the noncovered pension. In 1988, the WEP was modified so that the phaseout in the reduction to Social Security benefits applies to workers with 21 to 29 years of substantial covered employment, with the first replacement factor in the windfall formula adjusted on a sliding scale in 5% increments. The provision was included in the Technical and Miscellaneous Revenue Act of 1988 (P.L. 100-647).

31.

In 2004, S. 2455 was introduced by Sen. Kay Bailey Hutchison and H.R. 4391 was introduced by Rep. Kevin Brady to replace the current WEP with a proportional formula.

32.

For more information, see CRS In Focus IF11355, The Windfall Elimination Provision (WEP) in Social Security: Proposals for a New Proportional Formula, by Zhe Li.