EVENTS

Date:	Monday, June 19, 2006
Time:	12:00 PM -- 2:00 PM
Location:	Wohlstetter Conference Center, Twelfth Floor, AEI 1150 Seventeenth Street, N.W., Washington, D.C. 20036

June 2006

Social Security Reform: A Bipartisan Proposal

The Liebman-MacGuineas-Samwick Social Security plan was independently developed by former aides to President Bill Clinton, Senator John McCain (R-Ariz.), and President George W. Bush to demonstrate the types of compromises involved in a reform that might be acceptable to policymakers across the political spectrum. The plan would achieve sustainable solvency through progressive changes to taxes and benefits, and it introduces mandatory personal accounts and specifies important details that are often left unaddressed in other reform plans. The plan also illustrates that a compromise proposal can contain sensible but politically unpopular options--such as raising retirement ages or mandating that Social Security account balances be converted to annuities upon retirement--which could still realistically emerge from a bipartisan negotiating process but are often left out of reform proposals because of the political risk they present. The authors discussed their plan at a June 19 AEI panel.

Douglas Holtz-Eakin
Council on Foreign Relations

It is widely recognized that the fundamental entitlement program imbalances represent the most pressing challenge in U.S. fiscal policy.

Jeffrey Liebman
Harvard University

We wanted to find out whether a compromise on Social Security was possible. Most plans involve some combination of benefit cuts and tax increases, and most people believe that proposals should increase national savings. The substantive differences between Democrats and Republicans involve the mix between benefit cuts and tax increases and the extent to which we work within the current social security system, and these differences exist along a continuum. We found that it is difficult but possible to reach a deal.

There are plenty of sensible reform options, such as raising retirement ages or mandating that account balances be converted to annuities upon retirement, that may well emerge from a bipartisan compromise, but which are politically less attractive. We also tried to include details about how the disabled were treated, or about how annuities would work, so that they are not written down at 3 A.M. of whatever day a Social Security deal goes through Congress. We understand that our agreement does not need to face voters, and so our agreeing on a plan is a necessary condition for reform, but clearly not a sufficient condition.

Our plan has three main components:

• Traditional benefits are gradually reduced to the level affordable with a 12.4 percent payroll tax. This requires a reduction of benefits by about 40 percent compared with the levels specified in current law, but ensures that future benefits will be at least as large, in real dollars, as they are today.
• We introduce mandatory Personal Retirement Accounts (PRAs), which are equal to 3 percent of payroll and are funded half from trust fund revenue and half from new taxation. These accounts split the difference between the add-on and carve-out approaches.
• Rather than borrowing to finance the carve-out portion of the accounts, these costs will be covered by benefit cuts and by gradual increases in the earnings covered by Social Security tax.

Under our plan, Social Security surpluses would no longer be used to fund other government programs. The personal retirement accounts would be mandatory, but investment options would be limited to the fifteen or so investment plans that would be regulated by the government with full annuitization required. We would raise the earliest eligibility age from sixty-two to sixty-five, which would have positive labor market effects and protect people from myopically retiring too early. The plan would also protect the disabled and child beneficiaries from benefit cuts. We increase minimum benefits for low earners, increase the widow benefit, and reduce the spouse benefits for those married to high earners.

As a Democrat, I sought to find new ways to gain extra revenue and sought for the PRAs to not be a slippery slope to privatization. I believe that this plan achieves this.

Andrew Samwick
Dartmouth College

There were five areas of compromise in the plan, which would be necessary in the policy process:

• The balance between revenue increases and spending reductions. We split this difference 50/50.
• The combination of traditional benefits and PRAs. We keep traditional benefits to what can be afforded at 12.4 percent.
• We believe that 3 percent PRA contributions would build significant wealth relative to administrative cost, but would not overshadow traditional benefits.
• Incremental finances are locked into PRAs, which strengthens the link between the contributions and benefits. They would be inheritable, so when one spouse dies, the survivor could inherit the accrued benefits.
• Reductions of traditional benefits are greater for higher earners. Even the maximum earner would still get 50 percent of his or her current benefits.

Maya MacGuineas
New America Foundation

We sought to make the system solvent and sustainable. This would leave the system in balance over a seventy-five-year window. We feel that the plan is balanced and contains something for everyone to hate and love. We also try to balance the risks of the Social Security system. The existing system includes both financial and political risks which require diversification of investments.

As a political independent and deficit hawk, I care most about the fiscal responsibility of the system. We made transparent benefit and revenue changes. To pay the current promised benefits, the government would have to come up with $6.4 trillion. We have reduced this to $1.5 trillion. We also have the shortest transition period of any plan to pre-fund Social Security.

We believe that the plan would be likely to increase national savings, since accounts are a superior to trust funds as a place to store savings. This would increase savings but not cancel an increase in borrowing. We raised revenues in a non-distortionary way, since the PRA increase is less like a tax and more like mandatory saving. This is a generationally fair plan, which starts changes as soon as possible so that the burden is shared across generations.

We address the details with regard to disabilities and annuities. We think of practical but not politically popular ideas, such as raising the retirement age. PRAs are mandatory since it is too complex to craft offsets where people are allowed to opt out. The plan is transparent and without gimmicks.

Charles P. Blahous
National Economic Council

To answer key questions regarding a Social Security reform plan is to compare a plan with other plans. No reform plan will look good in comparison to current law, which proposes to pay benefits without charging for them.

As the baby-boom generation retires, spending is projected to soon soar above revenues. I asked how soon the plan would start to reduce and eliminate shortfalls. Plans without personal accounts reduce shortfalls earlier but never fully eliminate them. No plan ranks higher than this by both measures.

How well does the plan respond to the cost explosion? The extent to which plans rely on revenues, the more the burden is placed on the young, rather than controlling costs. This plan does not stand out from other plans with respect to cost containment. How transparent is the plan? It shifts less costs to the rest of the federal budget than any other I have seen.

There is no universal measure of benefit adequacy. There is some bipartisan consensus that low-wage earners should lose nothing in benefits. In its progressivity, the plan is broadly in the middle of the spectrum, depending on what return you get from PRA investment relative. However, due to its auxiliary benefits, it is slightly more progressive than others.

Under a voluntary PRA plan, you need to come up with a benefit offset to account for the fact that some people will opt out of paying in. A mandatory account would be excused from such complexities, but would not move future liabilities onto the current balance sheet in a way that can be controlled.

My main problem with the plan is that it severs the link between contributions and benefits for high earners by raising the cap for contributions, but not for benefits. This would put the political support of Social Security at risk.

I would put this plan slightly to the left of center, since it relies more on extra revenues than on cost containment. By objective measures, it is well-targeted to fiscal objectives and is transparent, but much depends on value judgments with respect to burden-sharing.

Jason Furman
New York University

The terms “taxes,” “benefits,” and “accounts” are slightly more malleable than they may seem at first. There is no clear distinction between a mandatory add-on account and a tax increase. Progressive redistribution through accounts is also possible by raising taxes and revenues from higher income people and putting it into the accounts of more moderate income people. Both of these are natural ways to reframe the issue and reach a compromise, and both are in this plan.

This plan achieves sustainable solvency without transfers from general revenues. We must ensure that the system is both sustainably and robustly solvent with respect to demographic shocks. This plan relies on net $1.1 trillion from the budget, but the trust fund is currently at $1.8 trillion, and so poses no adverse budgetary threat.

This plan increases national savings more than other plans, since it reduces benefits earlier and increases the add-on. This plan weighs more heavily on older generations.

Compared to this, I would raise retirement rates for moderate earners, raise the size of traditional benefits by removing the carve-outs, or capping accounts at $1800 to make them more progressive, pre-fund the accounts more, make the benefit formula more robust, and make the taxation more efficient.

How might policymakers mess this up? Voluntary accounts would reduce funding for myopic people--and those are the people we need to help most. They might partially annuitize or increase investment choices, allow people to dip into their accounts before maturity, or increase general revenue transfers.

Kent Smetters
AEI and The University of Pennsylvania

This is a sensible plan which does not rely on gimmicks, but I agree with Jason Furman that it is unlikely to be politically feasible. I am not sure that the plan is sustainably robust with respect to demographic shocks. I am not sure that mandatory annutisation is politically stable, and people might want to dip into these for college. This might be unfair to those who need to pay for cancer care and might erode the ideals of ownership and inheritability. Chile had a reasonable idea: that you had to annuitize enough to stay above poverty, but no more.

I agree with the add-on, since it gets new money into the system. When governments raise 1.5 percent PRA tax on the rich, they just reallocate their other income, whereas the poor must forego current income.

If an add-on is mandatory, we must call it a tax increase. If it is voluntary, then it is not a new tax. Is this a bipartisan plan? The Democrats will not buy the carve-out, which they will demagogue as privatization, and pretend that Social Security unbalances are not even a problem. The Republicans will like the problem, but the conservative base will see it as a tax increase.

The biggest issue for me is not private accounts, it is controlling the growth rate of benefits. Republicans must guarantee that all beneficiaries stay above the poverty line--but this is inflation-indexed rather than wage-indexed, and so will not expand unsustainably. Milton Friedman once said that Social Security started as an insurance program and evolved into a retirement program. I believe that the best solution is to move it back.

David Certner
AARP

There is much to be commended in this plan. There is an increase in the revenues, and it avoids additional borrowing. The plan does a version of a lockbox by putting funds into individual accounts, shielded from other government spending demands. However, as we have seen with the history of IRAs and 401(k) individual accounts, once you tell someone that something is his or her own money, it becomes very hard to lock it away from him or her for the long term.

We only need around 1.9 percent of payroll to get to solvency, but this plan includes benefit cuts of equivalent to 2.7 percent of payroll--dramatically more than you need to get to solvency. The cut in guaranteed benefits for the typical worker will be 40 percent more than is needed for solvency. We are doing this to accommodate the cut out of the payroll.

When you combine the benefits of the traditional plan with the PRAs, you see a benefit system that is less progressive than it is today. The low earners will get less and be forced to take on more risk, while the high earners will get more. We see the carve-out as a bad idea, regardless of its size. We should focus on the solvency of the system to maintain an adequate benefit structure. Carve-out accounts have little to do with solvency, efficiency, or national savings.

Over the past twenty-five years, we have seen pensions in this nation move from guaranteed defined-benefit to defined-contribution accounts where individuals take on more risk and responsibility--and we are yet to see whether this has worked, and it is a mystery why we might want to introduce it into the Social Security system.

This summary was prepared by Chris Pope, program manager of the National Research Initiative at AEI.