Mr. Chairman, Members of the Committee
I would like to applaud the Chairman and the Committee forholding these hearings today, and for your determination to goforward with trying to reform our nation's troubled retirementsystem. I hope that this means we are at last moving beyond thesterile and unproductive debate about whether Social Security isfacing a "crisis" or just a big problem.
Because whatever we call it, we cannot deny the fundamentalfacts. Social Security will begin to run a deficit in just 12years-that is, it will begin to spend more money on benefits thanit brings in through taxes. At that point, in order to continue topay promised benefits, it will have to draw on the Social SecurityTrust Fund. We have seen much debate about the Trust Fund recently,with some suggesting that it guarantees Social Security's solvencyuntil 2041, or even 2052. However, as Congressional Budget Officedirector Douglas Holtz-Eakin has noted "[The Trust Fund] has noreal economic resources….The key moments for Social Securityare in 2018. Cash-flow benefits will equal cash-flow payroll taxes,and then after that, the Social Security Administration will haveto come back to the rest of the budget for additional resources topay promised benefits."
Or as the Clinton Administration made clear in its FY2000budget:
"These Trust Fund balances are available to financefuture benefit payments…but only in a bookkeepingsense…. They do not consist of real economic assets that canbe drawn down in the future to fund benefits. Instead, they areclaims on the Treasury that, when redeemed, will have to befinanced by raising taxes, borrowing from the public, or reducingbenefits or other expenditures. The existence of Trust Fundbalances, therefore, does not by itself have any impact on thegovernment's ability to pay benefits."
This is not to say that the Federal government will default onthe bonds in the Trust Fund. I am not doubting the "full faith andcredit" of the U.S. government. However, that does not relieve theFederal government from the obligation to find the money with whichto redeem those bonds, currently $1.6 trillion in present valueterms. To put it in perspective, think of it this way. In 2018, thefirst year after Social Security begins running a deficit, theshortfall will be roughly as much as the Federal government spendson such programs as Head Start and the WIC program. The cost risesrapidly thereafter. By roughly 2023, the cost of redeeming enoughTrust Fund bonds to pay all the promised Social Security benefitswould be nearly as much as the cost of funding the Departments ofInterior, Commerce, Education, and the Environmental ProtectionAgency. By 2038, well before the theoretical exhaustion of theTrust Fund, you can add the Departments of Veterans Affairs,Energy, Housing and Urban Development, Justice, NASA, and theNational Science Foundation. Simply redeeming the Trust Fund willbegin to squeeze out all other domestic spending priorities.
Beyond 2042, once the Trust Fund is exhausted, the deteriorationin Social Security's finances only increases-and never gets anybetter. Overall, the present value of Social Security's unfundedobligations run to nearly $12.8 trillion (approximately $1.6trillion to redeem the Trust Fund, and $11.1 trillion in unfundedbenefits thereafter).
However, as troubling as these numbers may be, I believe thatthe debate over Social Security reform should not solely-or evenprimarily-be a discussion of solvency. Yes, solvency is important,and any responsible Social Security reform plan should restore theprogram to solvency, not just short-term actuarial solvency, butpermanent, sustainable solvency.
Still solvency is not enough. Instead, Social Security reformshould strive to build the best possible retirement system for ourchildren and our grandchildren. Thus, Social Security's currentsituation should not be seen as either a crisis or a problem, butas an opportunity to build a new and better program, based on thefundamental American values of ownership, inheritability, andchoice.
Under the current Social Security system you have no legal,contractual, or property rights to your benefits. What you getreceive from Social Security is entirely up to the 535 members ofCongress. But personal retirement accounts would give workersownership and control over their retirement funds. The money inyour account would belong to you-money the politicians (with alldue respect) could never take away. In short, they would own theirretirement.
Because you don't own you Social Security benefits, they are notinheritable. Millions of workers who die prematurely are not ableto pass anything on to their loved ones. But personal retirementaccounts would enable workers to build a nest egg of real,inheritable wealth.
Choice is part of the essence of America. Yet when it comes toretirement, Congress forces all Americans into a one-size-fits-all,cookie-cutter retirement program, a system that cannot pay thebenefits it has promised and in which we have no right to the moneywe pay in. With personal retirement accounts, workers who want toremain in traditional Social Security could do so. But youngerworkers who want a choice to save and invest for their retirementwould have that option.
With this goal in mind, not just to restore Social Security tosolvency, but to build a better retirement program that would giveworkers more ownership and control over their money, scholars atthe Cato Institute drew on our 25 years of experience studyingSocial Security, and developed a comprehensive proposal forcreating privately invested, personally owned accounts as part ofan overall reform of the Social Security system. This proposalbecame the basis for legislation introduced, on July 19, 2004, byyour colleague Rep. Johnson along with 18 originalco-sponsors.1 Rep.Johnson, together with Rep. Jeff Flake and 11 co-sponsors,reintroduced the bill in the 109th Congress, on January 21,2005.2
Under this proposal, workers under the age of 55 would have theoption of diverting their half of the Social Security payroll tax(6.2 percent of wages) to an individual account. The employer'sportion of the payroll tax would continue to be paid into theSocial Security system to provide survivors and disabilitybenefits, as well as to partially fund continuing benefits forthose already retired or nearing retirement. Workers choosing theindividual account option would forgo any future accrual of SocialSecurity retirement benefits. However, those workers who havealready paid into the current Social Security system, and thereforehave accrued benefits, would receive credit for those benefits inthe form of a recognition bond. This fully tradable bond would be azero coupon note maturing on the date of the recipient's normalretirement age.
Workers who do not choose the individual account option wouldcontinue to pay into and receive benefits from the current SocialSecurity system. However, for these workers, the initial SocialSecurity benefit formula will be adjusted to reflect price-indexingrather than the current wage-indexing. The result will be torestore Social Security benefits to a level payable with SocialSecurity's available revenue, while ensuring that future retireescontinue to receive the same level of benefits as those retiringtoday, on an inflation-adjusted basis. This change will be phasedin over a 35-year period, beginning in 2014.
This should not be seen as a benefit "cut." Indeed, benefitswill be higher in the future than they are today. While it is truethat future benefits would be less than what Social Securitypromises, such comparisons are meaningless because unless there isa substantial increase in taxes, the program cannot pay thepromised level of benefits.
That is not merely a matter of conjecture, but a matter of law.The Social Security Administration is legally authorized to issuebenefit checks only as long as there are sufficient funds availablein the Social Security Trust Fund to pay those benefits. Once thosefunds are exhausted, in 2041 by current estimates, Social Securitybenefits will automatically be reduced to a level payable withexisting tax revenues, approximately 73 percent of current benefitlevels.3
This, then, is the proper baseline to use when discussing SocialSecurity reform. Social Security must be restored to a sustainablelevel regardless of whether individual accounts are created.
As the Congressional Budget Office puts it:
A number of recent proposals to reform Social Securitycall for changes in the program's benefits. The effects of thoseproposals are frequently illustrated by comparing the new benefitsto those expected to arise under the policies put in place bycurrent law--showing whether they would be higher or lower and byhow much. However, because of scheduled changes in benefit rules, agrowing economy, and improvements in life expectancy, the benefitsprescribed under current law do not represent a stable baseline.Their value will vary significantly across future age cohorts.Thus, focusing on differences from current law will not fullyportray the effects of proposed benefit changes.4
I would also note that, although the Cato Plan and HR 350 applythe wage-index/price-index change to all income levels, if I wererewriting the proposal at this point, I would give very seriousconsideration to the blended approach advocated by Mr. Pozen. Doingso would refocus Social Security benefits on those who need itmost, and make the system more progressive.
The plan also called for establishing a new minimum SocialSecurity benefit equal to 100 percent of the poverty level,providing a significant increase over the current minimum benefit.I have attached the original Cato study setting out the details ofthe proposal and their rationale.
The plan has been scored by the Social Security Administration'sOffice of the Actuary (OACT), which concluded that it wouldeliminate Social Security's long-range actuarial deficit" and wouldrestore the system to permanent "sustainable solvency." I haveattached a study that the Cato Institute released last monthexploring OACT's findings in detail, as well as a copy of OACT'soriginal actuarial memo. However, to summarize, OACT foundthat:
- The "transition cost" (in present value) would be approximately$6.5 trillion. This is roughly half the $12.8 trillion unfundedliability of the current system. That is, the "6.2% Solution"ultimately saves taxpayers $6.3 trillion.
- The legislation also compares very favorably to other SocialSecurity reform plans. In terms of giving workers more control andownership of their retirement funds, the "6.2% Solution" clearlyprovides the most "bang for the buck."
- On a cash-flow basis, the legislation does require significantshort-term transfers of General Revenue. However, by 2046, thesystem would begin running surpluses, allowing any short-term debtto be repaid. Indeed, by the end of the 75-year actuarial window,the system would be running surpluses in excess of $1.8 trillion(in constant $2005)
- Much of the short-term cash-flow shortfalls are due to theredemption of recognition bonds, not to the diversion of payrolltaxes to the individual accounts. These recognition bonds conveymany benefits in terms of ownership as well as speeding the date atwhich Social Security changes from deficit to surplus. They areessentially a prepayment of future Social Security benefits, andnot a new expense. The Johnson-Flake bill is the only SocialSecurity reform bill with recognition bonds. The costs ofJohnson-Flake also include the cost of increasing the minimumSocial Security benefit to 100% of poverty, a significant increaseover the current minimum Social Security benefit.
- Individual accounts would eventually accumulate assets inexcess of $38 trillion (in constant $2005). That would lead tosubstantial new savings, new investment, and economic growth.
- Once short-term debt is paid off, the employer portion of thepayroll tax could be reduced to 3.04%. This would pay fordisability and survivors' benefits.
In short, the SSA analysis shows that Johnson-Flake can providelarge individual accounts while restoring Social Security topermanent sustainable solvency, and can do so in a fiscallyresponsible manner. While the up front costs will be significant,they will be less than for other big account plans, and eventuallythose costs will be more than offset by the savings to thesystem.
In addition, younger workers who chose the individual accountoption could receive retirement resources substantially higher thanwhat traditional Social Security can actually pay them.
Finally, Johnson-Flake gives workers ownership and control overtheir retirement income. It would give low- and middle-incomeworkers the opportunity to build a nest egg of real, inheritablewealth. It provides younger workers with greater choice. In short,if we measure a Social Security program not just as a matter ofdollars and cents, but as a matter of human liberty and individualdignity, Johnson-Flake provides a better way to take care of ourretirement.
1 HR 4895.
2 HR 530.
3 In practice, rather thanreduce each check sent to beneficiaries, the Social SecurityAdministration would stop sending out checks altogether until itaccumulates sufficient funds to pay "full" benefits. When thosefunds are exhausted, checks would again be withheld untilsufficient funds accumulate, leading to checks starting andstopping several times over the course of a year. The net effectwould be that total annual benefits would be reduced by the sameamount as if each month's benefits had been proportionallyreduced.
4 David Koitz, "MeasuringChanges to Social Security Benefits," CBO Long-Range Fiscal PolicyBrief no. 11, December 1, 2003.