Less than one year after President Bill Clinton pledged to "save Social Security first," he appears to have had second thoughts. Instead of proposing meaningful reforms to boost workers' retirement income and avert Social Security's coming financial crisis, he continues to broker an approach that merely shifts more IOUs into the program's trust fund. In addition to these accounting tricks, he is proposing a new entitlement plan called Retirement Savings Accounts. These downsized versions of last year's Universal Savings Accounts would use additional tax dollars to encourage people to save for their retirement. Both of these proposals allow the Administration to avoid making any hard decisions about the future of Social Security.

The old Universal Savings Accounts proposal was fundamentally poor, and the new RSA plan is little better. Although some lower- and middle-income Americans would be able to save more money for their retirement, these accounts would do little to increase savings for all workers; in fact, many of the workers who would benefit most from such accounts would not even be able to take full advantage of them. Moreover, the President's new tax-supported entitlement program would be cumbersome and expensive to manage. A better solution would be to allow workers to invest part of their existing payroll taxes in a safe portfolio of U.S. Treasury bonds and stock index funds.

How RSA Accounts Would Work.
Under the President's latest proposal, all single Americans who earn up to $40,000 a year and married couples earning under $80,000 a year who contribute to either an employer-sponsored 401(k) retirement plan or an RSA account would be eligible to receive matching contributions from their employer or a financial institution. Workers could receive a match on both their overall contribution and the first $100 they save. The level of match would depend on earned income. The employers and financial institutions would then be reimbursed by the government through a new tax credit equal to 100 percent of the savings match.

Flaws in the Plan.
Although workers could expect to earn much more for their retirement from money placed in RSA accounts than they now earn from their Social Security taxes, the flaws in the RSA plan would far outweigh this benefit:

1. RSA accounts would not save Social Security from insolvency.
   Instead of reforms that address Social Security's problems directly, the President recommends creating another program in addition to Social Security. Using part of the budget surplus to fund this program will not avert Social Security's impending financial crisis. The RSA program would spend more tax money and allow Washington to avoid making the painful and necessary decisions that would ensure Americans' retirement security. And Social Security would continue to face insolvency.

2. RSA accounts would become a new tax-financed entitlement.
   Dedicating promised surpluses to funding RSA accounts is a serious design flaw. If the federal government ever ran a deficit--something the Congressional Budget Office has admitted could occur--then working Americans would suffer. Congress would either have to end the RSA program or force future workers to pay higher taxes for that entitlement. Far from representing a tax cut, RSA accounts could become the first reason cited for future tax increases.

3. RSA accounts would be cumbersome and expensive to manage.
   According to the Clinton Administration, sometime after a worker filed an income tax return, the government would send a form certifying eligibility for some level of federal match on contributions to either an RSA account or the employer's 401(k) plan. The worker would make the contribution, and the matching amount would be paid by the financial institution or employer, who in turn would be reimbursed by the government.

Under the best of circumstances, this cumbersome process would be open to mistakes, misunderstanding, and fraud. The plan assumes that every employer or financial institution would participate and advance a matching dollar amount, expecting reimbursement by the government. In reality, participating employers or financial institutions would face compliance paperwork and endless bureaucratic delays. If fraud occurs, the risk would be borne by the employer or financial institution, not the federal government. Moreover, the actual costs of compliance are likely to be much higher than the $10 non-refundable tax credit offered by the plan.

4. RSA accounts are poorly designed for lower-income families.
   Most lower-income families--the very people who most need additional retirement savings--are unlikely to benefit from the program. Taxpayers who would have to choose between either paying the rent or medical bills or putting money aside for an event decades in the future are unlikely to have the extra dollars to save in these accounts and qualify for a federal match. As a result, the RSA program would evolve into just another middle-class entitlement.

5. RSA accounts would do little to increase savings.
   One reason many Americans cannot save enough for their retirement is that current taxes are too high. Leaving high Social Security taxes in place without reforming the system does nothing to help these Americans.

Real Reform Is Needed. The solution to these problems is not to create a new entitlement or savings program, but to turn part of the existing Social Security system into a true savings tool.  Congress should allow workers to invest a portion of their current Social Security retirement tax dollars in individually owned personal retirement accounts. The funds in these accounts could be put in a diversified portfolio of stocks, bonds, and other income-producing assets. Part of the budget surplus could be used to cover the short-term transition costs associated with this reform. Such personal retirement accounts, unlike RSA accounts, would enable Americans to enjoy higher incomes in retirement while placing Social Security on sound financial footing.

David C. John is Senior Policy Analyst for Social Security at The Heritage Foundation.