Close Nav

The Bipartisan Policy Center’s Tax-Heavy Social Security Plan


The Bipartisan Policy Center’s Tax-Heavy Social Security Plan

November 23, 2010


Click here for a printer-friendly version of this article

Social Security reform is hot again, as many of us knew that it soon must be. Delay continually makes the problem harder to solve, so the mere passage of time since President Bush’s reform effort has steadily lent greater urgency to the issue. In addition, Social Security’s finances took a further near-term hit from the recent recession. Add in the upcoming deadline for the President’s fiscal responsibility commission, as well as the passing of our biannual campaign “silly season,” and it is an opportune time for responsible people to step forward with solutions.

As a result, Social Security proposals are popping up seemingly everywhere. Last week I analyzed a proposal by fiscal commission co-chairs Simpson and Bowles. Commission member Representative Jan Schakowsky has also stepped forward with a plan (albeit one that only makes partial progress toward sustainable solvency) that I will analyze in a future article. Today I will describe the latest proposal from the Bipartisan Policy Center (BPC), an effort co-chaired by former CBO Director (Democrat) Alice Rivlin and former Republican Senator Pete Domenici.

Before I begin, one positive element of the BPC plan’s rollout is worth noting. The proposal is exceptionally well detailed and analyzed, even more so than Simpson-Bowles and much more so than the Schakowsky outline. BPC deserves great credit not only for developing a specific solution but also for providing a wealth of analytical information with it. So doing necessarily exposes them to more detailed critiques of their choices. Those of us who have disagreements with elements of the plan should recognize that the specificity put forward is a huge positive contribution to the debate.

Let’s start with the big picture first. The plan relies predominantly on additional tax revenue to shore up Social Security. The actuary’s analysis shows that over the next 75 years, roughly 89% of the improvement in the program’s actuarial balance would come from additional revenues, and only 11% from slowing the growth of system costs.

Perhaps the best way of visualizing this is with data taken directly from a chart in the Social Security actuary’s memo.


This graph compares projected cost growth under current benefit formulas with costs under the BPC plan. (Under literal current law, benefits would suddenly be cut in 2037 due to insolvency, but let’s put that aside for now and just focus on the benefits being promised under each approach). Everything here is expressed as a percentage of present-law taxable worker wages, so you can think of this as representing the share of each worker’s wage dollar absorbed by Social Security (It’s important to use the present-law tax base for each line to get an apples-to-apples cost comparison, since the plan would itself change the definition of wages subject to tax).

One huge policy challenge our nation faces is the explosion in system costs over the next quarter-century under current law. The BPC plan would somewhat slow this cost growth, but over the long term would actually result in an even more expensive system overall. Why this is so is best understood by looking at its individual provisions.

The biggest cost-saver under the proposal would be to index the program’s benefit formula (starting in 2023) for changes in life expectancy. This would slow the growth of benefits. This provision’s savings, however, are outweighed by others that would increase program obligations over the long term.

The proposal would gradually eliminate (from 2018-2028) the current exclusion from payroll taxation for employer-provided health benefits. This would increase Social Security tax revenue as well as benefit payments (which rise with the additional contributions). (This provision would also have very important effects beyond Social Security, reducing the projected growth of national health care costs).

Like Simpson-Bowles, the BPC plan would adopt the chained Consumer Price Index as the basis for future Social Security COLAs (as well as for other annual federal program and tax bracket adjustments). This is intended to more accurately model inflation, and produces savings by decreasing projected COLAs by roughly 0.3 points per year.

The BPC proposal would gradually increase the cap on wages subject to the Social Security tax from its current $106,800, phasing it up more rapidly than under current law to cover 90 percent of all national wages by 2049. The actuarial analysis does not quantify today’s wage equivalent for the eventual 2049 destination of the cap, but some recent estimates have been that if 90 percent of wages were subject to tax today, the cap would be somewhere around $180,000. This provision would also increase both revenues and benefit payments.

Other plan provisions would bring newly hired state and local workers under Social Security after 2020 (again increasing Social Security tax revenues and benefit obligations), subject cafeteria benefit plans to the payroll tax (also increasing revenues and obligations), and gradually reduce the rate of benefit accruals for high earners. Like Simpson-Bowles, the BPC plan would increase the minimum benefit for low-income workers as well as for the oldest beneficiaries. In addition, the analysis incorporates BPC’s broader tax reform proposals’ expected effects on Social Security finances.

As I explain in Social Security: the Unfinished Work, Social Security policy advocates sometimes have difficulty understanding one another’s positions when they have internalized different conceptions of what exactly the Social Security challenge is. Too often, we devolve into mutual accusations of ideological blinders or even malicious intent, when the real phenomenon is simply that different people define the problem in different ways.

Alice Rivlin and former Senator Domenici are among the public policy figures for whom I have the greatest respect. That said, I find myself in disagreement with many of the policy choices made in the BPC plan. In such a circumstance, it is often useful to reassess what one believes constitutes the Social Security policy challenge.

Simplifying considerably, I would rank the biggest problems facing Social Security somewhat like this:

1)The fiscal imbalance. The program’s scheduled benefits significantly exceed its projected revenues.

2)The cost explosion. Without changes, the program will impose tax burdens on younger generations far higher than previous ones were asked to shoulder.

3)Misleading accounting. Current Trust Fund accounting masks the true size of costs being passed to future generations.

4)Pay-as-you-go financing. Because Social Security is a pure income transfer rather than a savings program, it can’t increase total retirement saving by increasing benefits; instead, one generation’s benefits are merely subtracted from the next generation’s potential retirement saving. Moreover, pay-go financing imposes worsening treatment on younger generations as society ages.

5)Work disincentives. The program still incorporates 1935 inducements both for spouses and younger seniors to drop out of the workforce, when today we sorely need these Americans’ productivity.

6)Imprecise benefit targeting. Though the program is progressive on average, overall income redistribution is relatively haphazard, reducing program efficacy as a bulwark against poverty.

The nature of their proposal suggests that the BPC sees Social Security’s challenges somewhat differently. The proposal would indeed close much of the fiscal imbalance (the first criterion), though less than Simpson-Bowles. (Where Simpson-Bowles would eliminate 100% of annual deficits by the 75th year, the BPC proposal would eliminate about 88% by then, and less afterward). Some of the plan provisions would also make incremental improvements in targeting benefit payments on those most in need, the sixth criterion listed above.

The BPC proposal however does not really attempt to solve problems #2 through #5. On criterion #4 (lack of saving) and #5 (work incentives), this is fairly understandable. The federal government’s current fiscal condition makes it difficult now to finance even partial pre-funding of future program obligations. Moreover, even in 2005 when our fiscal condition was much better, many political figures unfortunately adopted strong ideological positions against any pre-funding. This is bad for younger generations, but it may not be politically plausible to do anything about it right now. Also, BPC’s isn’t the only proposal to leave Social Security’s flawed work incentives uncorrected.

More concerning, however, is the plan’s lack of a solution for problem #2 (rising costs) or #3 (Trust Fund accounting). The plan basically leaves the program’s projected long-term cost explosion on the books; it makes decent headway on the cost problem over the next quarter-century but later actually adds to the long-term cost burdens.

It also relies more than similar proposals on Trust Fund accounting conventions. In the later years, Social Security would still be running significant and growing annual deficits; “solvency” would then be preserved in part by counting on future general taxpayers to make steadily increasing cash payments of interest from general revenues to the Social Security Trust Fund.

The BPC proposal is a serious plan, backed up with an impressive amount of analysis and documentation. It’s also a plan under which Social Security, just as under current law, would grow much more expensive to taxpayers in the future than it is now.

Charles Blahous serves as one of the two public trustees for the Social Security and Medicare programs. He is also the author of Social Security: The Unfinished Work.

e21 Partnership

Stay on top of the issues that matter to you most

By clicking subscribe, you agree to the terms of use as outlined in our Privacy Policy.



Main Error Mesage Here
More detailed message would go here to provide context for the user and how to proceed
Main Error Mesage Here
More detailed message would go here to provide context for the user and how to proceed