Michael Tanner’s Testimony on Social Security Reform
before the House Ways and Means Committee


I very much appreciate the privilege of appearing before youtoday, and the opportunity to discuss how Social Security reformcan benefit vulnerable populations. It is now generallyacknowledged that Social Security is facing severe future financingproblems. The program will begin running deficits in just 12 years,and is facing total unfunded obligations of roughly $12.8 trillion(including the cost of redeeming the Trust Fund). As a result,changes in the program are inevitable.

In making these changes, however, it is particularly importantthat we consider their impact on the most vulnerable Americans whodisproportionately depend on Social Security. For example, thepoorest 20 percent of Americans receive nearly all of theirretirement income from Social Security, while the wealthiest fifthof Americans receive less than 20 percent of their retirementincome from the system. It is also important to understand that itis not just reform that will affect these vulnerable Americans, butso to will a failure to reform the system. Since Social Securitycurrently cannot pay promised benefits, those benefits willeventually have to be reduced by roughly 26 percent, a reductionthat will fall heaviest on those who can least afford it.

On the other hand, reform, properly structured, can not onlyprotect the poor and vulnerable from these otherwise inevitablebenefit cuts, but can actually produce an improved Social Securitysystem that will leave them better off. We can give low incomeworkers a chance to build real inheritable wealth. We can give theman ownership stake in the American economy. And, while maintaininga safety net, we can give them a chance to earn a higher rate ofreturn, leading to higher retirement benefits that would liftmillions of seniors out of poverty.

Social Security has elements of both an insurance and a welfareprogram. It is, in effect, both a retirement and an anti-povertyprogram. However, in attempting to combine these two functions, ithas ended up doing neither particularly well. While much time hasbeen spent discussing Social Security's shortcomings as aretirement program, far less attention has been paid to itsinadequacies as an anti-poverty program.

There is no question that the poverty rate among the elderly hasdeclined dramatically in the last half century. As recently as1959, the poverty rate for seniors was 35.2 percent, more thandouble the 17 percent poverty rate for the general adultpopulation. Today, it has declined to approximately around 10percent.

Clearly Social Security has had a significant impact on thistrend. Studies suggest that in the absence of Social Securitybenefits more than half of seniors would have income below thepoverty level. This suggests that receipt of Social Securitybenefits lifted millions of seniors out of poverty. Moreover, thepercentage of elderly in poverty after receiving Social Securitybenefits has been steadily declining in recent years, indicatingthe increased importance of Social Security as an anti-povertyremedy.

However, there is a superficiality to this line of analysis. Itassumes that any loss of Social Security benefits would not beoffset through other sources of income. In other words, it simplytakes a retiree's current income and subtracts Social Securitybenefits to discover, no surprise, that the total income is nowlower and, indeed, frequently low enough to throw the retiree intopoverty.

That much should be obvious. Social Security benefits are asubstantial component of most retirees' income. It constitutes morethan 90 percent of retirement income for one-quarter of theelderly. Nearly half of retirees receive at least half of theirincome from Social Security. The question, therefore, is notwhether the sudden elimination of Social Security income wouldleave retirees worse off--clearly it would--but whether in theabsence of Social Security (or an alternative mandatory savingsprogram) retirees would have changed their behavior to provideother sources of income for their own retirement.

However, even taking the idea of Social Security as ananti-poverty tool on its own terms, the evidence suggests that thecurrent Social Security is inadequate . After all, despitereceiving Social Security benefits, roughly one out of ten seniorsstill lives in poverty. In fact, the poverty rate among seniorsremains slightly higher than that for the adult population as awhole. And, among some subgroups the problem is far worse. For thepoverty rate is over 20 percent among elderly women who are nevermarried or widowed and roughly 30 percent among divorced orseparated women. African-American seniors are alsodisproportionately left in poverty. Nearly a third ofAfrican-Americans over the age of 65 have incomes below the povertylevel.

In addition, lifetime Social Security benefits depend, in part,on longevity. As a result, people with identical earnings historieswill receive different levels of benefits depending on how longthey live. Individuals who live to be 100 receive far more inbenefits than individuals who die at 66. Therefore, those groups inour society with shorter life expectancies, such as the poor andAfrican-Americans, are put at a severe disadvantage.

Of course, Social Security does have a progressive benefitformula, whereby low-income individuals receive proportionatelyhigher benefits per dollar paid into the system than do high-incomeworkers. The question, therefore is to what degree shorter lifeexpectancies offset this progressivity.

Using income as the sole criteria, the literature is mixed. Somestudies, such as those by Eugene Steuerle and Jan Bakja of theUrban Institute and Dean Leimer of the Social SecurityAdministration conclude that shorter life expectancies diminish butdo not completely offset Social Security's progressivity. However,there is a growing body of literature, including studies by DanielGarrett of Stanford University, the RAND corporation, JeffreyLiebman, and others that show the progressive benefit formula iscompletely offset, resulting in redistribution from poor people towealthy.

The question of Social Security's unfairness to ethnicminorities appears more straightforward, particularly in the caseof African-Americans. At all income levels and all ages,African-Americans have shorter life expectancies than do whites. Asa result, a black man or woman, earning exactly the same lifetimewages, and paying exactly the same lifetime Social Security taxes,as his or her white counterpart, will likely receive far less inlifetime Social Security benefits.

This disparity has a significant impact on the concentration ofwealth in our society. Social Security benefits are notinheritable. A worker can pay Social Security taxes for 30 or 40years, but if that worker dies without children under the age of 18or a spouse over the age of 65, none of the money paid into thesystem is passed on to his heirs. As Cato Senior Fellow JagadedeshGokhale, has noted, Social Security essentially forces low-incomeworkers to annuitize their wealth, preventing them from making abequest of that wealth to their heirs.

Moreover, because this forced annuitization applies to a largerportion of the wealth of low income workers than high incomeworkers, it turns inheritance into a "disequalizing force," leadingto greater inequality of wealth in America. The wealthy are able tobequeath their wealth to their heirs, while the poor cannot.Indeed, Gokhale and Boston University economist Laurence Kotlikoffestimate that Social Security doubles the share of wealth owned bythe richest one percent of Americans.

Martin Feldstein of Harvard University reaches a similarconclusion. Feldstein suggests that low-income workers substitute"Social Security wealth" in the form of promised future SocialSecurity benefits for other forms of savings. As a result, agreater proportion of a high-income worker's wealth is in fungibleassets. Since fungible wealth is inheritable, while Social Securitywealth is not, this has led to a stable concentration of fungiblewealth among a small proportion of the population. Feldstein's worksuggests that the concentration of wealth in the United Stateswould be reduced by as much as half if low-income workers were ableto substitute real wealth for Social Security wealth.

Properly constructed, a Social Security reform plan includingpersonal accounts can solve these problems. HR 530, introduced byyour colleague Mr. Johnson, provides an excellent example of howthis would work. Mr. Johnson's bill would allow younger workers tosave and invest their half of the Social Security payroll tax (6.2percent of wages) through personal accounts. Because workers wouldown the money in their accounts-which they do not under the currentsystem-that money would be fully inheritable. If they die beforeretirement prematurely, they would be able to pass all the money intheir account on to their loved ones; death after retirement wouldstill leave substantial unused portions for their heirs.

And, it is not just future generations who would benefit fromthis ownership. Personal accounts would give low-income workers achance to build a nest egg of real wealth for the first time intheir lives, giving them a real and personal stake in the economy.As Michael Sherraden of Washington University in St. Louis hasshown, ownership can have significant beneficial impact on avariety of social pathologies, not only increasing work effort andthe propensity to save, but even reducing crime, drug abuse, schooldrop out rates, and illegitimacy. Giving people an ownership stakein America-something that HR 530, with its recognition bonds doeseven more than other personal account plans-could be considered oneof the most important anti-poverty proposals we couldundertake.

Finally, HR 530 would also establish an enhanced safety net toprotect the most vulnerable. It leaves current survivor anddisability benefits unchanged. However, it also includes a newminimum Social Security benefit equal to 100 percent of the povertylevel, a significant increase over today. Thus, under Mr. Johnson'sbill, no eligible senior would ever again retire into poverty.

Other personal account plans, including Mr. Ryan's and to alesser extent Mr. Shaw's, would also represent a significant boostfor low income and otherwise vulnerable Americans.

In summation, Social Security reform is inevitable. If we simplyfall back on the old ways of raising taxes and cutting benefits, wewill significantly harm those most at need. If we do nothing, weend up with a benefit reduction that the poor and vulnerable canill afford. However, by making personal accounts part of any SocialSecurity reform, we can give low-income workers a chance to build anest egg of real inheritable wealth. In combination with anenhanced safety net, we can provide vulnerable workers with a newand better Social Security system.

Thank you.