From Discourse Magazine <discoursemagazine@substack.com>
Subject Brookings’ Constructive Social Security Proposal
Date January 17, 2025 11:03 AM
  Links have been removed from this email. Learn more in the FAQ.
  Links have been removed from this email. Learn more in the FAQ.
View this post on the web at [link removed]

The U.S. Social Security system has been sinking into deepening trouble of late, its finances heading towards collapse, and with fewer friends in positions of power willing to do anything about it. There was once a time (the 1970s and 1980s) when each party’s leadership cared enough about Social Security to join in making politically difficult decisions to preserve its solvency. Unfortunately, a political schism emerged in the 2000s and 2010s, during which time only fiscal conservatives remained willing to sound the alarm, while progressives took to denying the reality of Social Security’s worsening condition. More recently, even purported conservatives have become increasingly unwilling to step forward and call for solutions.
But on January 3, the Social Security Administration’s chief actuary released an analysis [ [link removed] ], requested by Rep. Steny Hoyer (D-Md.), of a comprehensive reform proposal developed by Wendell Primus of the Brookings Institution. Not only is the proposal a constructive one, but it may also have the potential to jump-start a desperately needed discussion of how best to rescue Social Security from impending insolvency.
Elsewhere, I have written about the Social Security financing crisis [ [link removed] ] and why legislators must act to address it. I am also on record [ [link removed] ] as recommending that Social Security financing reforms focus on cost containment [ [link removed] ] rather than raising taxes. The Brookings proposal contains a mixture of both approaches but relies predominantly on revenue (that is, tax) increases. Although many of its provisions differ from policy recommendations I have made, the proposal is potentially very useful in that it would preserve the most valuable features of Social Security through specific measures to increase revenues and (to a lesser extent) cut costs, all of which are based on solid rationales.
The Brookings Proposal Would Preserve Social Security’s Most Valuable Features
First, the proposal charts a better direction than we are currently heading. Ideally, lawmakers would have long been debating the right blend of cost containment measures and revenue increases to preserve Social Security solvency, but that’s not the world we inhabit. Instead, legislators have appeared content to allow Social Security to become insolvent and to react only belatedly to that crisis by bailing out the program with trillions in additional debt financed by general revenue. This would be the worst of all potential outcomes.
For fiscal conservatives, a general revenue bailout would be an obvious disaster because it would add enormously to the public debt and destroy the last vestiges of financial discipline under current law, wherein Social Security spending is constrained by the amounts worker contributions can fund. But it would equally be a disaster for progressives because it would terminate the program feature that Social Security recipients have paid for their benefits and derived income security from that. Thereafter, Social Security would be just another welfare program competing for resources within the government’s general fund.
While I don’t favor every tax increase in the Brookings proposal, its blend of revenue increases with some cost containment is far preferable to simply giving the program an unlimited claim on federal debt issuance and destroying its historical design as a contributory insurance benefit. If the Brookings proposal reminds legislators on the left as well as the right that Social Security as Americans know it is worth preserving, this by itself would be a great contribution.
Second, while it would be better if we could still fix Social Security painlessly without additional revenues, that is no longer possible. In past years, one could argue that sustainable solvency could be achieved simply by reindexing the growth of future benefits, without tax increases and without benefit reductions in real (inflation-adjusted) terms. Unfortunately, due to policymakers’ inaction, that’s no longer where we are [ [link removed] ]. Unless lawmakers are willing to countenance real Social Security benefit declines going forward, the program will need additional revenues.
Third, many of the provisions in the Brookings proposal are meritorious in their own right. Agree or disagree with them, there is a sensible policy rationale underlying each one. These provisions are worth consideration separate and apart from their quantitative effects on program financing.
Fourth, the Brookings proposal avoids many problems in other Social Security plans that rely on collecting more revenues. A frequent flaw of such proposals is that they wouldn’t actually fix the problem but simply kick the can down the road, leave the program on an unsustainable track and require still more tax increases at a future date. The Brookings plan avoids that pitfall. It would not only achieve actuarial balance on average, but it would maintain balance into the more distant future, meaning that under current projections lawmakers wouldn’t need to follow up with additional tax increases later on.
While the proposal as a whole is constructive, some of its provisions are more attractive than others. An optimal solution would do more to constrain the system’s explosive cost growth and rely less on raising taxes.
Key Provisions of the Bookings Proposal
Transfer all revenue from Social Security benefit taxation from Medicare to Social Security, including retroactively. Because this provision would reallocate revenues from benefit taxation (which currently contributes funding to both Social Security and Medicare) that are already collected, it doesn’t represent a net improvement in the federal government’s ability to finance its spending. Still, there is a defensible logic to the proposed change; it makes intuitive sense that the income from taxing Social Security benefits should belong to Social Security rather than being split between Social Security and Medicare, as it is now. Moreover, Social Security’s financing shortfall is much larger than the one facing Medicare Hospital Insurance; in other words, Social Security needs it more.
A change I’d recommend to this provision is to eliminate its retroactivity. Perhaps the policy should have been different in the past, but it wasn’t, and we can’t collect taxes from long ago in the future. The adverse fiscal effect of a retroactive transfer would be a large additional commitment of general-revenue-financed debt to Medicare. (My recommended change to this provision would need to be offset with additional savings within Social Security.)
Gradually increase the Normal Retirement Age (NRA) for participants in the top 40% of earnings. Brookings is to be applauded for including an eligibility age change in the plan, despite the fact that political opportunists persistently sow confusion about its effects. A rising NRA actually doesn’t delay the age at which Americans may claim benefits; that’s determined instead by the Early Eligibility Age, currently 62. The NRA simply changes the age at which one can receive full benefits—or, put another way, it changes the benefit amount for a given claim age.
Adjusting eligibility ages is essential: Continuing failure to do so can only result in workers facing perpetually higher tax rates as longevity lengthens, or alternatively, lower levels of annual retirement income. The only way to avoid this Hobson’s choice is to get a handle on the rising number of years during which beneficiaries collect Social Security payments. Brookings’ proposal would gradually increase the NRA to 70, but only for the top 20% by income and not until 2054. Everyone in the bottom 60% would have the same NRA as under current law, and the NRA for people in the 60-80th percentile range would be somewhere in the middle.
A change I’d recommend to this provision would be to apply the full proposed NRA adjustment to everyone. There’s no compelling policy reason to introduce the complexity of different eligibility ages for different incomes, especially since any differential effects of the change on lower-income workers (who tend to have shorter lives) could be offset by progressive changes to the benefit formula. Workers should know their eligibility age without having to consult their relative position on the national earnings distribution. (My recommended change to this provision would achieve additional savings.)
Change the number of computation years in the benefit formula from 35 to 40. Social Security computes benefits by tracking a worker’s earnings each year, translating each of those past years of earnings into today’s equivalents (by indexing them for national wage growth), identifying the worker’s highest 35 earnings years after that indexing, averaging them (to create each worker’s Average Indexed Monthly Earnings), and applying a benefit formula to that average. It’s a complicated process, but the essence of it is that a worker’s benefit is based on the average of their top 35 earnings years.
This is a problem for a number of reasons, one being that once people have worked for 35 years, they continue to pay payroll taxes without accruing much in the way of benefits. Americans have actually figured this out [ [link removed] ], dropping out of the labor force right when Social Security starts punishing them for continuing to work. Increasing the number of computation years from 35 to 40 would save money and strengthen system finances because then the worker’s total earnings would be divided by 40 rather than by 35 to calculate benefits, meaning individual benefit payouts would be smaller. But a positive from workers’ perspective is that it would preserve rewards for an additional five years of their employment—specifically, years 36 through 40.
An adjustment I’d recommend to this provision would be to include all earnings years in the numerator of the benefit formula, even if the denominator is fixed at 40. That way, no matter how many years one works, one would always be rewarded with additional benefits for doing so. (My recommended change to this provision would need to be offset with additional savings.)
Increase the amount of earnings to which the Social Security payroll tax is applied. Cost containment [ [link removed] ] is preferable to increasing taxes for a number of reasons. However, any bipartisan Social Security solution will undoubtedly include some revenue increases, and surveys often show that increasing the amount of earnings subject to the payroll tax has as wide support as any other proposal. The Brookings proposal takes a prudent approach of raising the cap on payroll-taxable earnings to cover 90% of all national employment earnings—the percentage it covered at the time [ [link removed] ] of the last Social Security financing fix (it’s about 83% today). Importantly (unlike some proposals), it would maintain the connection between workers’ contributions and their benefits. This link between individual contributions and benefits is at the heart of Social Security’s unique success as a social insurance program, and Brookings deserves credit for recognizing this.
Enact various policy reforms that would increase net immigration. Legal immigration is a boon to Social Security finances because it increases the number of payroll tax-paying workers. Technically, illegal immigration also bolsters Social Security finances because many who are illegally present pay payroll taxes under invalid Social Security numbers while never claiming benefits. Illegal immigration creates a plethora of other problems, however, so reform should promote legal immigration only, the higher-skilled the better.
The Brookings proposal would increase legal employment-based immigration, create new immigration pathways for health and personal care aides by designating them as shortage occupations, and also implement key provisions of the Dignity Act that would allow undocumented “Dreamers” to pursue permanent resident status. My expectation is that something less than these proposals would be enacted in any bipartisan immigration reform, and thus, while such reforms could be positive for Social Security finances, they would be less so than indicated in the proposal’s actuarial memorandum.
Apply federal income tax to all Social Security benefits paid to individuals with incomes over $100,000, and joint filers over $125,000, with these thresholds indexed for price inflation. Under current law, at most 85% [ [link removed] ] of an individual’s Social Security benefits are subject to income tax. The reason for that limit is that Americans’ payroll taxes were already subjected to the income tax when they were first paid, and in some cases, the amount already taxed equals as much as 15% of their Social Security benefit.
Thus, if we begin taxing all Social Security benefits of some high-income individuals, as Brookings proposes, those affected could justifiably claim that as much as 15% of their benefits are being subject to double taxation. But under current law, baby boomers and Gen Xers will take more out [ [link removed] ] of Social Security than they put in, and it’s virtually impossible to craft a fair solvency solution that exempts these generations from making any net contribution. Since legislators will not want to reduce benefits for those already collecting them, exposing more of high-income recipients’ benefits to income taxation may be a more palatable way of having these generations contribute at least something to the solution.
Apply the Social Security payroll tax to distributions to pass-through business owners. Social Security gets most of its financing from its payroll tax. Social Security creates a connection between individual contributions and benefits by imposing a flat payroll tax rate on a worker’s earnings, while also calculating benefits as a function of those same taxable earnings. There are four basic ways to increase this revenue: (1) Increase the payroll tax rate; (2) increase the amount of worker earnings subject to the tax; (3) broaden the forms of compensation to which the payroll tax applies; and (4) tax something else. Proposals in category 4 are usually bad ones because they would destroy the vital link between individual contributions and benefits that is the heart of Social Security.
The Brookings proposal would do a little of each of the first three, and this provision represents approach 3. Broadening the payroll tax base makes sense in principle; however, many possible ways of doing so, such as applying the payroll tax to compensation in the form of health benefits, risk hitting lower-income workers disproportionately hard. If, instead, the Brookings provision preserves the link between earnings and benefits, is progressive and reduces distortions of compensation practices, it is worth considering as a preferable way to increase payroll tax revenue.
Gradually phase out the dependent spouse benefit. Despite being well intended, Social Security’s nonworking spouse benefit is one of its most problematic features in practice. It is both regressive and punishes employment. Under current law, an individual can receive a much larger spousal benefit simply by marrying a high-income person than the benefit a low-income worker can earn through a lifetime of payroll tax contributions. Reforming the benefit makes sense, but rather than phase it out altogether, lawmakers might consider capping it on the high-income end, so that no one receives more as a nonworking spouse than a minimum-wage earner can accrue over a full career of payroll tax contributions. (My recommended change to this provision would need to be offset with additional savings.)
Increase the Social Security payroll tax from 12.4% to 12.6%. Responses to surveys usually indicate Americans’ preference for raising the cap on taxable wages over raising the payroll tax rate. However, this conceptual preference sometimes disintegrates when policymakers attempt to flesh out actual proposals. Broadening the payroll tax base and raising the tax cap both feature an unpalatable choice between either breaking the contribution-benefit link (bad policy) or obligating additional payouts based on the increased earnings subject to tax (inefficient). Raising the tax rate avoids either problem. The Brookings proposal envisions a relatively slight rate increase from 12.4% to 12.6%. Although an ideal proposal would maximize cost containment and minimize additional revenue needs, it is reasonable to fulfill part of a given revenue target via a slight adjustment to the tax rate.
Cover all newly hired state and local government employees, and replace the previous Windfall Elimination Provision (WEP) and Government Pension Offset (GPO) with new provisions that use the worker’s total (covered and noncovered) earnings history to achieve parity between Social Security participants with state pensions and those without. Some state and local government employees do not contribute to Social Security, and there have long been proposals to cover them. These proposals were seen as problematic because they’d essentially amount to a revenue grab by Social Security, further destabilizing state and local pension plans that are already underfunded as it is.
The recent enactment of the misnamed Social Security Fairness Act [ [link removed] ], however, effectively grants state and local employees Social Security benefit windfalls that they didn’t pay for [ [link removed] ]. In this updated context, it is more appropriate for these state and local employees to begin contributing to Social Security. The Brookings proposal would also replace the previous WEP and GPO with reformed versions that would more accurately achieve parity between Social Security participants with comparable earnings levels but different employment histories.
General Recommendations
Taken together, my recommended modifications to the Brookings proposal’s provisions would lessen the improvements in Social Security’s actuarial status. Offsetting savings would be needed for that reason alone. However, there’s another reason additional cost savings would be needed.
There is one large problem facing Social Security that the Brookings proposal as currently drafted wouldn’t solve. Under current law, Social Security costs are projected to grow relative to GDP by roughly 23% from now through the 2070s. That’s a problem because it translates into rising burdens on employment and economic production. The Brookings proposal [ [link removed] ] as a whole doesn’t address this—in fact, it would exacerbate the problem both in the near term and the long run, in the sense that system costs would be even higher than currently projected.
More needs to be done to slow the growth of Social Security costs, even before considering the federal government’s problem of how to finance the even faster growth of Medicare and Medicaid. How much a legislated solution should slow the growth of Social Security costs is a subjective value judgment, but even if lawmakers decide it’s reasonable to address just half of its projected growth relative to GDP, financing the rest with tax increases, significantly more cost savings will be needed.
Fortunately, there are plenty of prudent ways to slow system costs beyond the provisions in the Brookings plan. Among them are the following:
Changing the benefit formula so that higher earners accrue benefits at a slower rate than they do under current law.
Reindexing the growth of the benefit formula so that average benefits grow more slowly, thus moderating the program’s cost growth. This could be done by tying the rate of benefit growth to price inflation rather than average wage growth (which tends to grow faster than inflation) as it is under current law.
Making technical corrections, such as adopting a more accurate (chained) Consumer Price Index or tying the benefit formula to workers’ annual earnings rather than lifetime average earnings. (This latter measure would save money by preventing the system from continuing to treat high-income sporadic workers as though they are low-income workers in need of extra assistance. In addition to strengthening system finances, this reform would better target system resources.)
Any or all of these measures could be added to several of the features of the Brookings proposal to create a plan that is more balanced between cost containment and increased revenues.
Regardless of whether readers or policymakers agree with these subjective recommendations, the Brookings proposal is a constructive Social Security plan at a time when we desperately need one. Unless lawmakers across the ideological spectrum engage on the question of how to preserve Social Security as a solvent, self-financing program, it cannot maintain that design. Abandoning the self-financing nature of the program would be a tremendous loss for all Americans, from progressives to conservatives. The Brookings proposal could be a useful starting point for a discussion of how best to preserve Social Security in the form Americans have come to know and depend on.

Unsubscribe [link removed]?
Screenshot of the email generated on import

Message Analysis