Executive Summary

This study examines the fiscal implications of a proposal to add paid parental leave benefits to the Social Security system. The proposal would allow workers to claim early Social Security benefits in the event of the birth or adoption of a child. In turn, the workers would delay their retirement by a period commensurate with the cost of their leave.

This study finds that this new paid leave program would worsen Social Security’s already troubled financial outlook. Specifically:

• The program would cost $10.5 billion in 2019 and $227.6 billion from 2019 to 2034, when the Social Security Trust Funds are projected to reach depletion;
• $226.2 billion (99.4 percent) of the program’s cost would be a net cost to the Social Security Trust Funds prior to the Trust Funds’ exhaustion;
• The $226.2 billion in net costs to the Social Security Trust Funds would advance the exhaustion date by about six months.

Since the Treasury already borrows to fulfill its Social Security Trust Funds’ obligations, the $226.2 billion in net program costs would be financed by government debt.

Introduction

The Independent Women’s Forum’s (IWF) recent proposal to provide paid parental leave through Social Security has gained considerable interest from policymakers, policy experts, and paid leave advocates. Under the proposal, workers would be able to claim up to 12 weeks of early Social Security benefits in the event of the birth or adoption of a child. In return, the workers would delay their retirement for a period commensurate with the cost of their leave benefits. The retirement deferral is intended to offset the cost of the paid parental leave benefits, so that over a lifetime there is no impact on the Social Security Trust Funds or rise in government spending.

Given that Social Security is already projected to be insolvent in 2034, however, there are serious concerns about the long-term viability of this proposal and its impact on retirement benefits. Social Security already runs a cash deficit and its Trust Funds are expected to reach full exhaustion in 2034. This study estimates the annual budgetary cost of the paid leave proposal and its impact on the combined Social Security Trust Funds. It finds that in 2019 the paid parental leave program would provide $10.5 billion in benefits to 3.6 million workers. Between 2019 and the Trust Funds’ depletion in 2034, it estimates program expenditures would total $227.6 billion. $226.2 billion of those benefits (99.4 percent) would go to workers who will reach retirement age after the Trust Funds’ exhaustion. Consequently, the paid leave program would add new benefit payments of $226.2 billion from the Social Security Trust Funds prior to the Trust Funds’ exhaustion, as the vast majority of benefit recipients between 2019 and 2034 will retire after the Trust Funds’ depletion. As a result, the parental leave program would accelerate total Trust Funds’ exhaustion by roughly six months.

The Current Landscape of Paid Family Leave Proposals

The IWF parental leave proposal is the latest in a series of proposals, as both liberal and conservative policymakers search for ways to provide the benefit without overburdening the federal budget, taxpayers, and employers.[1]

In the United States, only one federal law is currently in place to guarantee family and medical leave. Under the Family and Medical Leave Act of 1993 (FMLA), certain workers are allowed up to 12 weeks of unpaid, job-protected leave for the birth or adoption of a child (“parental leave”), to care for an ill family member (“family care leave”), or to recover from one’s own serious illness (“medical leave”).[2] Yet, no federal laws guarantee workers are paid during their time away from work. In fact, the United States is the only developed country that does not guarantee workers some form of paid leave.[3]

In recent years, a series of federal paid leave proposals have surfaced from across the political spectrum. Many on the left prefer Senator Kirsten Gillibrand’s FAMILY Act, which would create a new social insurance program that distributes 12 weeks of payroll tax-financed leave benefits to workers who require time off to care for a new child, to care for an ill family member, or to recover from their own serious medical condition.[4] On the right, meanwhile, there have been myriad proposals. Senator Deb Fischer’s proposal to offer tax credits to businesses that provide paid leave became the first-ever paid family leave program in U.S. history when it became law at the end of 2017 with the enactment of the Tax Cuts and Jobs Act (TCJA).[5]^{, [6]} President Trump proposed to provide six weeks of paid parental leave through the unemployment insurance system.[7] Representative Martha Roby’s Working Families Flexibility Act would allow workers to accrue paid time off for working overtime in lieu of additional pay.[8] Representative John Katko’s Working Parents Flexibility Act would allow employers to offer workers the option to divert a portion of their pretax earnings into a parental leave savings account like a 401(k)-retirement savings account.[9] Finally, Representative Mimi Walters’s Workflex in the 21st Century Act would exempt employers that provide sufficient paid time off and workplace flexibility from local and state paid leave mandates.[10]

Policy groups have also introduced a handful of paid leave proposals. In particular, the American Enterprise Institute (AEI)-Brookings Institution working group on paid family leave (of which AAF’s Ben Gitis and Douglas Holtz-Eakin are members) proposed a social insurance program that would universally provide eight weeks of paid parental leave financed by payroll taxes and cuts in existing spending and tax expenditures.[11] Additionally, AAF has offered two iterations of a proposal (one in a joint report with AEI) to provide paid leave benefits specifically to low-income workers.[12]^{, [13]}

IWF’s Proposal: Incorporate Paid Parental Leave into Social Security

IWF’s proposal would provide paid leave with Social Security benefits in a way that is intended to be self-financing and would not require a new payroll tax.[14] Workers would have the option to collect up to 12 weeks of Social Security retirement benefits early in their careers after the birth or adoption of a child. In return, workers would pay for their own benefit by delaying retirement for a period long enough to offset the cost of their paid leave. While the details of the required deferral period have yet to be determined, the delay would be short relative to the length of a working career. IWF suggests that workers would need to defer retirement by six weeks to offset the cost of 12 weeks of paid leave. More recent estimates from the Urban Institute indicate that the deferral period would be closer to 25 weeks.[15]

The paid parental leave benefit would be available to both women and men. The value of the benefit payments would be based on the same progressive benefits formula used to calculate Social Security Disability Insurance (SSDI) benefits. Workers would receive benefits equal to 90 percent of their first $895 of average monthly earnings, 32 percent of all average earnings from $895 to $5,397 per month, and 15 percent of all earnings above $5,397 per month. All workers would be subject to a maximum benefit of $2,788 per month. Like SSDI, the benefits would also be based on average earnings to date. To be eligible for the benefit, new parents would be required to have worked at least two of the four quarters leading up to the birth or adoption of the child and a total of at least four quarters so far in their lifetime. By definition, all year-round workers would be eligible for the benefit.

The Current State of the Social Security System

In theory, this proposal is intuitive and logical—the government simply advances the money to beneficiaries—but it is vital to consider the current state of Social Security to determine how it will work in practice. The most recent report from the Trustees of Social Security’s Old-Age, Survivors, and Disability Insurance (OASDI) programs shows that the nation’s primary safety net for retirees, survivors, and the disabled remains in financial distress and that, absent reform, the program will fail to meet its promises to future seniors.^[16]

The report estimated that the combined (retirement and disability) Social Security Trust Funds will be bankrupt by 2034. The Trustees’ report provides additional metrics that make clear the program’s structural imbalances. In 2017, Social Security spent $952.5 billion but only collected $911.5 billion in non-interest income, leaving a cash deficit of $41.0 billion. 2017 was the eighth year in a row that Social Security ran a cash deficit, with the program running a cumulative deficit of $457 billion since 2010. The Treasury funds these deficits by borrowing from the public—in effect, raising the overall debt issuance by the federal budget.

According to the Trustees, over a 75-year horizon Social Security’s promised benefits exceed projected payroll taxes and Trust Funds redemptions by $13.2 trillion; this “unfunded liability” is $700 billion higher than was estimated last year. Social Security faces an imbalance as a share of taxable payroll of 2.84 percent, the second highest reported imbalance since 1982, which was the last report issued by the Trustees before Social Security was last meaningfully reformed. According to the Trustees, the combined OASDI Trust Funds will be exhausted by 2034, or 16 years from now, which is the shortest horizon to exhaustion since 1982.

The Trustees report paints a distressed picture of Social Security’s financial health and proves that the present course is unsustainable. Social Security is now contributing to the annual deficit, while promised benefits vastly exceed planned funding. The implications of failing to reform the status quo are a 21 percent scheduled reduction in benefits in 2034. After the projected exhaustion of the Social Security Trust Funds, Social Security revenue will fund only 79 percent of promised benefits. This reduction deteriorates further, to 74 percent, by 2092. Absent reform, to meet promised benefits over the long-term, Congress would have to increase payroll taxes immediately by 31 percent, from a rate of 12.4 percent to 16.27 percent.

Social Security’s troubled outlook undermines the long-term viability of the IWF paid parental leave proposal. When the Social Security Trust Funds reach exhaustion in 2034, the parental leave benefit would face the same across-the-board cuts as retirement benefits, and workers would receive only 79 percent of their promised payments. Moreover, the paid parental leave program would also worsen Social Security’s outlook. In the following, we estimate the budgetary cost of the IWF paid parental leave proposal and its impact on the Social Security Trust Funds.

Methodology

Budgetary Cost

To estimate the budgetary cost of the IWF paid parental leave proposal, this study assumes that program participation would mirror private sector leave-taking patterns seen under the FMLA, the primary federal family and medical leave policy currently in effect. In particular, it assumes that take-up and duration of leave under the program would match those of employed people who took parental leave and are eligible for 12 weeks of job protection under the FMLA. The FMLA, however, does not guarantee those workers are paid. Thus, the resulting estimate from this analysis could be a lower bound, because the introduction of a paid leave benefit would likely increase take-up and average duration.

Data on take-up and duration under the FMLA come from a 2012 survey on the FMLA conducted by Abt Associates on behalf of the Department of Labor.[17] Using that data, this study estimates the portion of FMLA-eligible workers who went on parental leave in 2012 and the length of their leave. Additionally, it uses data from the Current Population Survey (CPS) March 2017 Annual Social and Economic Supplement to analyze the distribution of participating workers and their earnings, which it in turn uses to estimate the program benefits.[18] The analysis excludes anyone 67 and over from eligibility, as they would be eligible for retirement benefits and would not claim parental leave benefits.[19]

Since the CPS provides data on the workforce for one year, this study is unable to use it to take into account fully the proposal’s eligibility requirement of working at least four quarters total and at least two of the last four quarters leading up to a birth or adoption. Instead this study restricts eligibility to all year-round workers (roughly 80 percent of employed people), which the requirements inherently include.[20] Since it is possible to meet the requirement without being a year-round worker, this assumption likely results in understating the costs. Additionally, one-year of CPS data do not enable an adequate calculation of the average lifetime wages, from which benefits would be based. As a result, this analysis bases benefits on the previous year’s earnings. This factor could result in the estimate overstating the cost of the leave program, since wages are commonly lower early in careers than in the year just prior to taking leave and would thus result in a lower calculation of lifetime average wages and the resulting benefits. These factors likely offset each other somewhat. Yet, without better information, this study is unable to determine their net effect on the cost estimate.

Impact on the Trust Funds

The addition of this leave benefit would result in significant upfront costs as new parents claim the benefit early in their careers, while all the offsetting savings occur at the back end when they delay retirement decades later. The challenge with this concept is that if enacted now, the increased upfront costs and the offsetting future savings essentially straddle the exhaustion of the Social Security Trust Funds. In the immediate term, the benefits claimed would widen the cash deficit and further draw down the Trust Funds. Since the Treasury already meets its Trust Funds’ obligations by borrowing, these new paid leave benefits also would be financed with additional debt. Later on, however, since very few leave-takers would reach retirement age before the Trust Funds’ depletion in 2034 (just 16 years from now), most would not have the opportunity to neutralize their impact on the Trust Funds prior to their exhaustion when delaying their retirement. With significant additional benefits and few offsetting savings between now and 2034, the paid leave program would accelerate the exhaustion of the Social Security Trust Funds.

This study estimates the impact on the Social Security Trust Funds’ exhaustion date in two separate steps. First, it estimates the total budgetary cost between 2019 and 2034 (depletion period) and the portion of that cost that would be offset by benefit recipients who will retire before depletion (and delay collection of their Social Security retirement benefits). The study assumes that the retirement deferral period accurately offsets the cost of individuals’ paid leave. Thus, the paid leave benefits that would go to workers who reach retirement age before 2034 would be fully offset before the Trust Funds’ exhaustion and have no impact on its financial state. The entirety of the benefits that go to workers who reach retirement age after 2034, however, would be financed as a cost on the Trust Funds.

To estimate the portion of benefits that would go to workers who reach retirement age before 2034, this study again turns to the 2012 FMLA data. Using that data, this study analyzes the age distribution of workers who took parental leave. Assuming the same age distribution of new parents each year going forward, it estimates the portion of benefit recipients who in each year from 2019 to 2034 would be old enough to retire before 2034. Using the same methods employed to estimate the overall annual cost of the program, it estimates the total portion of the benefits that would be provided to these older workers, which would be offset with delayed retirement before 2034 and not negatively impact the Trust Funds. Reductions in the Social Security Trust Funds, however, would finance the cost of paid parental leave benefits for all younger workers each year from 2019 to 2034.

Second, to project the actual impact on the Trust Funds’ depletion date, this study adds the new costs to projected benefit payments under the Trustees’ intermediate cost scenario and estimates the effects of these additional costs on future Trust Funds’ balances. In general, given that the combined OASDI program is running a cash deficit, the additional costs reduce net interest income over the period 2019 to 2034.

Results

This study finds overall that from 2019 to 2034 the IWF paid parental leave program would be costly, almost entirely financed by reductions in the Social Security Trust Funds (and, consequently, by federal borrowing), and would accelerate the Trust Funds’ depletion date. Since existing Social Security retirement outlays are already so large, however, the impact of the new paid parental leave program on accelerating the Trust Funds’ exhaustion would be relatively modest.

Cost of Paid Leave Proposal

We estimate that in 2019, the IWF paid parental leave program would provide a total of $10.5 billion in benefits to 3.6 million workers, or 2.4 percent of employed people.

Table 1: 2019 Participation, Duration, Weekly Benefits, and Total Benefits Paid

Participants	3,600,000
Participation as Percent of Employment	2.4%
Average Duration (weeks)	7.1
Average Weekly Benefit	$404
Total Benefits Paid (millions)	$10,500

Weekly benefits would average $404, or 37 percent of average weekly earnings. Additionally, workers would claim the leave benefit for an average of 7.1 weeks.

At $10.5 billion, the cost the IWF proposal is of similar magnitude to the cost of the AEI-Brookings working group proposal, which comes out to $11.8 billion in 2019 under similar methods.[21] Yet, the IWF proposal’s cost is much smaller than the more expansive FAMILY Act, which would cost $85.9 billion in its first year.[22]

Over time, the cost of the program would likely grow with population growth and inflation. Consequently, between 2019 and 2034, benefit outlays for paid parental leave would cumulatively add up to hundreds of billions of dollars. Table 2 contains the total projected cost of the IWF proposal from 2019 to 2034 and the portions of the spending that would go to workers who reach retirement age either before or after 2034.

Table 2: Total Program Benefits and Portions Paid to Workers Who Retire Either Before or After Social Security Trust Funds’ Exhaustion, 2019-2034

Worker Category	Benefits (Millions)	Benefits (%)
All Workers[23]	$227,600	100.0%
Retiring Before 2034 (Trust Funds Outlays Offset)	$1,400	0.6%
Retiring After 2034 (Added Costs to Trust Funds)	$226,200	99.4%

From 2019 to 2034, the Social Security Trust Funds would provide $227.6 billion in paid parental leave benefits. Of those benefits, however, only $1.4 billion or 0.6 percent would be offset by workers who reach retirement age before 2034. The remaining $226.2 billion or 99.4 percent would go to workers who reach retirement age after 2034. Consequently, the overwhelming majority of program benefits would add costs to the Social Security Trust Funds. Moreover, the Treasury meets its Social Security Trust Funds’ obligations by borrowing and adding to the federal debt. Thus, between 2019 and 2034, 99.4 percent of the IWF paid parental leave program would be financed by federal borrowing.

Impact on the Trust Funds’ Depletion

Accordingly, the upfront costs of the paid leave program will essentially accelerate the Social Security Trust Funds’ exhaustion date. Based on the costs estimated in this analysis, however, it will not substantially accelerate that date. Rather, it would likely advance the exhaustion date by about six months, with this estimate subject to a great deal of uncertainty. This accelerated exhaustion is driven by reductions in net interest income from the higher upfront costs, which leaves the Trust Funds with reduced assets than would have otherwise been the case. From 2019 to 2034, the leave proposal would add $226.2 billion in costs to Social Security prior to the Trust Funds’ exhaustion. To the extent that Social Security is in a cash flow deficit, all new costs would reduce net interest income to the Trust Funds, reductions which would compound over time.

Yet, the new costs are relatively small compared to the magnitudes of the existing flows in and out of the Trust Funds. For instance, while the paid parental leave program would add $226.2 billion in costs from 2019 to 2034, Social Security provided over $950 billion in benefits in 2017 alone. Thus, the diminished interest income effect is not sufficient to substantially accelerate Social Security’s insolvency. Given the trajectory of Social Security and the implications of the Trust Funds’ exhaustion, however, any advancement of that date is concerning. Moreover, it would be unwise to avoid needed reforms that would ensure the Social Security’s solvency.

Issues Regarding Actuarial Neutrality

The Social Security Trust Funds’ projected depletion in 2034 complicates the issue of actuarial neutrality. Although the paid leave program is intended to be actuarily neutral, and this analysis assumes actuarial neutrality, the automatic cuts to benefits after 2034 means that the paid leave benefits, as outlined by IWF, would not be actuarily neutral. In particular, those who claim paid leave benefits prior to the Trust Funds’ exhaustion but retire after the exhaustion would receive full benefits for paid leave but only 79 percent of their promised benefits for retirement. Nevertheless, policymakers have options for ensuring actuarial neutrality. In particular, the paid leave benefit could be based on accrued benefits instead of promised benefits (i.e., 79 percent of promised benefits). Additionally, policymakers could further extend the delay in retirement to offset the actuary imbalance.

Conclusion

Perhaps the trickiest aspect in paid leave policy is finding a solution that meaningfully expands access to paid leave without overburdening the public with higher taxes, fewer benefits elsewhere, or additional debt. The concept of the IWF paid parental leave proposal is appealing due to its intended self-financing nature. Those who would collect paid leave benefits from Social Security would effectively pay for their own leave by delaying retirement and reducing their future Social Security benefits. In theory, this structure is logical, as it implies no new taxes and no new spending.

The proposal, however, would not work so well in practice because of Social Security’s troubled financial outlook. Social Security already runs a cash deficit and its Trust Funds are projected to go bankrupt by 2034. Under the IWF proposal, the vast majority of workers who claim paid leave benefits would retire after 2034 and would not offset the cost of their leave until Social Security has already gone bankrupt. Consequently, the paid parental leave benefit would add $226.2 billion in new costs to the Social Security Trust Funds prior to exhaustion (financed by federal borrowing) and modestly advance the Trust Funds’ depletion.

[1] Kristin A. Shapiro, “A Budget-Neutral Approach to Parental Leave,” Independent Women’s Forum, January 9, 2018, http://iwf.org/publications/2805496/Policy-Focus:-A-Budget-Neutral-Approach-to-Parental-Leave.

[2] “Family and Medical Leave Act of 1993,” H.R.1, 103rd Congress, 1993, https://www.congress.gov/bill/103rd-congress/house-bill/1.

[3] Gretchen Livingston, “Among 41 Nations, U.S. Is the Outlier When It Comes to Paid Parental Leave,” Pew Research Center, September 26, 2016, http://www.pewresearch.org/fact-tank/2016/09/26/u-s-lacks-mandated-paid-parental-leave/.

[4] “Family and Medical Insurance Leave Act,” S.337, 115th Congress, 2017, https://www.congress.gov/bill/115th-congress/senate-bill/337.

[5] “Strong Families Act,” S.1716, 115th Congress, 2017, https://www.congress.gov/bill/115th-congress/senate-bill/1716.

[6] “Tax Cuts and Jobs Act,” Conference Report, H.R.1, 115th Congress, 2017, https://files.taxfoundation.org/20171215175456/TCJA_Conference_Report.pdf.

[7] Ben Gitis, “The Costs and Distribution of Benefits of the President’s Paid Leave Proposal,” American Action Forum, American Action Forum, May 23, 2017, https://www.americanactionforum.org/research/costs-distribution-benefits-presidents-paid-leave-proposal/.

[8] “Working Families Flexibility Act of 2017,” H.R.1180, 115th Congress, 2017, https://www.congress.gov/bill/115th-congress/house-bill/1180/text.

[9] “Working Parents Flexibility Act of 2017,” H.R.2533, 115th Congress, 2017, https://www.congress.gov/bill/115th-congress/house-bill/2533.

[10] “Workflex in the 21st Century Act,” H.R.4219, 115th Congress, 2017, https://www.congress.gov/bill/115th-congress/house-bill/4219.

[11] “Paid Family and Medical Leave: An Issue Whose Time Has Come,” Working Group Report, American Enterprise Institute & Brookings Institution, June 6, 2017, http://www.aei.org/publication/paid-family-and-medical-leave-an-issue-whose-time-has-come/.

[12] Ben Gitis, “The Earned Income Leave Benefit: Rethinking Paid Family Leave for Low-Income Workers,” American Action Forum, August 15, 2016, https://www.americanactionforum.org/solution/earned-income-leave-benefit-rethinking-paid-family-leave-low-income-workers/.

[13] Ben Gitis and Angela Rachidi, “Affordable and Targeted: How Paid Parental Leave in the US Could Work,” American Action Forum and American Enterprise Institute, March 29, 2017, https://www.americanactionforum.org/solution/affordable-targeted-paid-parental-leave-us-work/.

[14] Kristin A. Shapiro, “A Budget-Neutral Approach to Parental Leave,” Independent Women’s Forum, January 9, 2018, http://iwf.org/publications/2805496/Policy-Focus:-A-Budget-Neutral-Approach-to-Parental-Leave.

[15] Melissa M. Favreault and Richard W. Johnson, “Paying for Parental Leave with Future Social Security Benefits,” Urban Institute, April 19, 2018, https://www.urban.org/research/publication/paying-parental-leave-future-social-security-benefits.

[16] Douglas Holtz-Eakin, Tara O’Neill Hayes, and Jonathan Keisling, “The Future of America’s Entitlements: What You Need to Know About the Medicare and Social Security Trustees Reports,” American Action Forum, June 5, 2018, https://www.americanactionforum.org/research/the-future-of-americas-entitlements-what-you-need-to-know-about-the-medicare-and-social-security-trustees-reports/.

[17] “Family and Medical Leave in 2012: Public Use File – For Employees,” Abt Associates/Department of Labor, 2014, https://www.dol.gov/whd/fmla/survey/.

[18] “Current Population Survey, 2017 Annual Social and Economic (ASEC) Supplement,” Census Bureau and Bureau of Labor Statistics, March 2017, accessed at National Bureau of Economic Research, http://www.nber.org/data/current-population-survey-data.html.

[19] Contact the authors for more detail on the methodology used to estimate the cost of the program.

[20] “Table 1. Work experience of the population during the year by sex and extent of employment, 2015-2016,” Bureau of Labor Statistics, December 5, 2017, https://www.bls.gov/news.release/work.t01.htm.

[21] It is important to note that the estimated $11.8 billion cost of the AEI-Brookings proposal differs slightly from the previous AAF cost estimate of $11.1 billion. That is because in the previous report, AAF used a slightly different methodology based on Census data on birth rates among employed women, whereas this report utilizes parental leave patterns in the FMLA data. Since the two methods yield similar estimates, they are likely comparable. For the previous analysis of the AEI-Brookings proposal, see Ben Gitis and Douglas Holtz-Eakin, “Analysis of AEI-Brookings Working Group Proposal on Paid Parental Leave,” American Action Forum, June 7, 2017, https://www.americanactionforum.org/research/analysis-aei-brookings-working-group-proposal-paid-parental-leave/.

[22] Ben Gitis, “The Earned Income Leave Benefit: Rethinking Paid Family Leave for Low-Income Workers,” American Action Forum, August 15, 2016, https://www.americanactionforum.org/solution/earned-income-leave-benefit-rethinking-paid-family-leave-low-income-workers/.

[23] Assumes program spending would grow with nominal gross domestic product, as projected by the Congressional Budget Office.