PROPOSED
"SAVINGS INCENTIVES” WOULD CAUSE REVENUE HEMORRHAGE IN FUTURE DECADES
Proposal Also Would Heavily Benefit Wealthy Taxpayers, While Weakening Pension
Coverage
for Workers and Shifting Costs to Future Generations
By
Robert Greenstein and
Joel Friedman
| PDF of
full report HTM of fact sheet PDF of fact sheet |
| If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader. |
On January 31, the Administration unveiled a package of dramatic proposals that ultimately would shield from taxation a large share of all income earned on savings and investments. Despite the profound changes the proposal would make, the President’s budget shows that this proposal would raise revenues by $15 billion through 2008.
Embedded in the proposal are budget gimmicks that temporarily raise revenues over the next few years. Revenue losses would begin later in the decade, and would continue to grow substantially for a long period of time, with the proposal bleeding the budget heavily in future decades. The ultimate budget cost could be massive, with much of the damage occurring in the same period that the baby boomer generation will be retiring in large numbers. The nation already faces the prospect of budget deficits of unmanageable proportions in these years.
While the proposal is being touted as a way to increase national saving and to help workers save for their retirement, in fact the proposal may have the opposite effects. National saving only rises if there is a net increase in the combination of private saving (by individuals and institutions) and public saving (by government when it runs a surplus). This proposal would reduce public saving over time because it increases the deficit, and it is unlikely to generate enough new private saving to offset this decline in public saving. The proposal’s incentives for encouraging new savings are weak because they primarily benefit high-income individuals, who are more likely to shift existing savings from taxable accounts to the new tax-preferred vehicles rather than undertake new savings in response to the tax break. Further, the proposal would likely lead to an erosion of employer-sponsored pensions for ordinary workers, if business owners and executives felt they could save enough for their own retirement through these new savings vehicles and therefore chose to forgo the cost of offering retirement plans through the workplace.
This analysis explores the proposal and finds it would have the following effects.
- Reduce Federal Revenues Over the Long Run — The proposal would cause a large "double-hit” on federal revenues in future decades, swelling deficits that the Congressional Budget Office, the General Accounting Office, and independent analysts already project will rise to alarming levels during those years. First, by eliminating income tax on a steadily growing share of all income earned on savings and investments, the plan would depress the revenue the government collects. Second, the gimmicks contained in the plan would shift substantial amounts of revenue from future decades into the next five years, allowing the proposal to raise revenue in the five-year budget window that Congress will use this year, but then punching a still-deeper hole in future budgets.
|
High Cost of Administration’s Savings Proposals
A major analysis of the savings proposals by the
The reports finds that some aspects of the Administration's savings proposals, such as ending contributions to traditional IRAs and encouraging holders of traditional IRAs to roll over their funds into the new Retirement Savings Accounts, would raise revenue in the short run. But this temporary revenue increase simply represents a timing shift and corresponds to an equal or greater revenue loss in the future. Moreover, these future revenue losses would be occurring at a time when the budget situation will be much worse than it is now, due to the pressures created by the retirement of the baby boom generation. The report finds that the short-term revenue gains amount to "mortgaging future revenues ... at very unfavorable terms for the government."
__________________________ |
- Hurt States by Reducing Their Revenues and Raising Borrowing Costs — The proposal also would cause state governments to lose substantial amounts of revenue over time. States generally tie their definition of taxable income to the federal definition; they conform to federal rules regarding Individual Retirement Accounts and similar tax-code features. As a result, the revenue hemorrhage that the proposal would cause would affect state governments as well. Moreover, the proposal may reduce the appeal of tax-free state and local government bonds, since affluent investors could secure tax-free earnings through the retirement account and savings account tax shelters the plan would make available to them. This could force states to offer somewhat higher interest rates on the bonds they issue in order to attract sufficient investment in the bonds, which would increase state costs. Since states must balance their budgets each year, states would have to cut services for their residents or raise state taxes in order to offset the losses in state revenues and the increases in state expenditures that the proposal would engender.
- Provide Windfalls to Wealthy Taxpayers — The proposal would confer windfalls of rather massive proportions on the nation’s wealthiest individuals. A wealthy couple with two children would be able to put $45,000 a year into tax-sheltered saving vehicles on which all earnings would be tax free. (Moreover, this figure does not include amounts the couple could deposit into tax advantaged, employer-based retirement accounts. After 2006, such a couple where both parents are working would be able to put an additional $30,000 a year into such accounts, for a total of $75,000 a year.) Over time, wealthy individuals could shift substantial amounts of their savings and investments into these tax-favored accounts, with the interest, dividends and capital gains income earned on the amounts accruing tax free. The result would be of enormous benefit to wealthy individuals who have large amounts of assets they can shift into these accounts. As such individuals shift growing amounts of assets into the accounts, federal and state governments would incur mounting revenue losses. Further, these savings proposals, coupled with the Administration’s proposal to permanently repeal the estate tax, would allow wealthy individuals to secure large amounts of investment income tax free and then pass it on to their heirs tax free. No tax would ever be paid on these investment earnings.
- Moves Toward Consumption Tax — The plan can be viewed as taking a large step toward converting the current progressive income tax into a less-progressive consumption tax. Past proposals to convert the income tax to a consumption tax generally have included new consumption taxes — such as a value-added tax — to replace the lost income tax revenues. The Administration’s plan eliminates over time much of the income tax now collected on income earned on savings and investments, thereby moving the income tax toward being a consumption tax, but would do so without replacing the lost income tax revenue with any new consumption tax revenue. It is that step that produces the long-term revenue hemorrhage.
- Creates Tax Shelter Opportunities — Moreover, the plan could create opportunities for abusive tax shelters that a pure consumption tax would not have. Individuals could use home equity loans to borrow funds, deposit the borrowed funds in the tax-free savings accounts and escape tax on all income earned on the borrowed funds, and still take tax deductions on the interest payments they make on the borrowed funds. Such abusive tax shelters would add to federal and state revenue losses.
- Likely to Reduce National Saving — While the proposal is promoted as increasing national saving, it is more likely to reduce national saving than to increase it, thereby slowing long-term economic growth. National saving equals the sum of private saving — i.e., saving by private individuals and institutions — and public saving, which consists of government surpluses. (If governments run deficits, public saving is negative and reduces overall national saving.) It is national saving, not private saving, that affects the pool of capital available for investment and thereby influences economic growth.
The proposal would cause large reductions in public saving over time because it would swell the deficit. And while it might lead to increases in private saving, any such increases likely would be modest. Low-, moderate-, and middle-income families tend to spend most or all of their incomes each year. While the proposal could induce some of these families to save more, most such families are limited in how much more they can afford to save. Indeed, even Treasury Department documents accompanying the budget acknowledged that "one-third of all Americans have no assets available for investment, and another fifth have only negligible assets.”[1] Thus, over half of Americans have insufficient assets to take advantage of these savings proposals.
Those who would be able to take greatest advantage of the expanded IRA and other savings accounts that the proposal would create are wealthy individuals who can afford to place very large sums in the generous new tax-sheltered accounts.
Economic research indicates that when affluent individuals use savings vehicles of this nature, they largely shift savings that they already have from investments that are subject to the income tax to vehicles that are exempt from tax. When a taxpayer shifts savings from one type of account to another in this manner to avoid taxes, it adds nothing to the total amount of savings in the nation. Indeed, the New York Times has reported that the plan has been greeted with widespread skepticism from economists, who are dubious the plan will do much to boost private saving (see box).[2]
Since the reduction in public saving that would occur as a result of the increase in the budget deficit is likely to be greater than the increase in private savings, the overall effect on national saving — and long-term economic growth — is more likely to be negative than positive. Because of the proposal’s effect on the deficit, it is not plausible to argue that the proposal will significantly boost economic growth.
- Likely To Reduce Pension Coverage for Ordinary Workers — The proposal is likely to lead to a reduction in pension coverage for ordinary workers. Today, if a business owner or executive wants to put more than $6,000 a year into tax-advantaged retirement or saving accounts for himself or herself and a spouse, the owner or executive must offer a pension plan through his or her firm, and such a plan must cover the firm’s employees as well as the owners and executives. (The $6,000 figure is scheduled to rise to $10,000 by 2008.)
Under the Administration’s new proposal, by contrast, business owners and executives could put away $30,000 a year for themselves and their spouses through expanded IRAs and the new "Lifetime Savings Accounts,” plus an additional $7,500 a year for each child they have, without having to offer any retirement plan whatsoever through their firm. Owners could boost their own tax-sheltered savings by using the new tax-free savings vehicles. They could then decline to offer employer-based coverage through their firms, thereby saving money for the firms and improving their bottom lines. Whether to offer retirement plans is often a close call for small and medium-size businesses. The expansion and creation of these retirement and saving accounts is likely to lead over time to fewer firms offering pension coverage for their workers and making pension contributions on their workers’ behalf.
Indeed, Brian Graff, executive director of the American Society of Pension Actuaries, told the New York Times last week, "If you’re a small-business employee, what this could potentially mean is your employer will no longer offer a program for you.” Graff observed that since small-business owners and their spouses would be able to put $30,000 each year in the new tax-advantaged accounts, they would have little need to create tax-advantaged savings plan at work. "It’s the rank and file who are going to lose out of the opportunity to get a matching contribution [from their employer],” he observed.[3]
As a recent Wall Street Journal article explained, the proposal also dilutes the provisions of current law that prevent owners and executives from directing an overwhelming share of a firm’s retirement contributions to themselves while doing little for their workers.[4] The weakening of these protections would enable many firms that retain retirement plans to skew the plans more heavily toward owners and executives and scale back contributions for ordinary workers.
For all of these reasons, the plan is likely to be disadvantageous to ordinary Americans over time. They would benefit from not having to pay tax on amounts they deposit in the new tax-sheltered saving accounts that the proposal would establish. But they would be harmed in multiple ways: from the federal budget deficits that the plan would enlarge and the slower economic growth, higher long-term interest rates, and/or budget cuts in programs that benefit them that ultimately would ensue; from the state budget cuts and/or tax increases that many states would have to institute to make up for the revenue losses they would incur; and from the reductions in pension coverage for some workers.
In his State of the Union address, President Bush said he would not push problems on to future Congresses, future Presidents, and future generations. A White House aide in a prior Administration once tellingly confided, "Watch what we do, not what we say.” In fact, few proposals in recent American history would do more to push problems on to future policymakers and future generations than this proposal is likely to do.
The remainder of this analysis explains in more detail how the proposal would work and why it would have these effects.
IRAs and Employer-sponsored Retirement Plans Under Current Law
Individuals whose employers offer a retirement plan can make contributions to retirement accounts. Their employers can make contributions to these accounts, as well. Under most employer retirement plans, the funds contributed to such retirement accounts by the employee and the employer are not counted as part of the employee’s income for income tax purposes, and the earnings that accrue in the accounts accumulate free of tax. When individuals retire and withdraw funds from these accounts, the amounts withdrawn count as taxable income.
Traditional IRAs operate on the same principle as these types of employer-sponsored retirement plans. Taxpayers eligible for a traditional IRA may deposit up to $3,000 a year in such an IRA.[5] Their spouses may deposit $3,000 as well. The amounts that a taxpayer deposits in these IRAs are deducted from the taxpayer’s income for income tax purposes, and the amounts earned in the IRA accounts accumulate tax free. When funds are withdrawn from the accounts after retirement, the amounts withdrawn count as taxable income.
Roth IRAs, developed in the 1990s, are different. Under Roth IRAs, deposits made into the IRA accounts are not tax deductible. Once funds are deposited, however, all amounts earned on the accounts — and all withdrawals in retirement years — are tax free.
The maximum amounts that may be contributed to a Roth IRA and a traditional IRA are the same (currently $3,000). In addition, the two types of IRAs offer similar size tax benefits and have similar costs over time. But the timing of the tax benefits and of the cost to the federal government are different. Under traditional IRAs, the bulk of the cost occurs upfront when tax deductions are taken as the deposits are made. Under Roth IRAs, the major costs occur later, when withdrawals are made tax-free during retirement. Roth IRAs are sometimes referred to as "backloaded IRAs” because the costs to the government are backloaded and tend to occur primarily in years outside the five-or ten-year period that Congressional budgets typically cover.
The Administration’s Savings Proposal
The Administration’s proposal contains a number of elements, establishing three new savings vehicles. First, it proposes to eliminate existing IRAs and replace them with Retirement Savings Accounts. The RSAs would operate in virtually the same manner as Roth IRAs, under which there is no upfront deduction for contributions, but earnings grow tax free and no tax is paid on funds when they are withdrawn. Second, the Administration would create new Lifetime Savings Accounts. The LSAs would operate like RSAs, but funds could be used for any purpose rather than just for retirement and funds could be withdrawn at any time. Finally, the Administration proposes to create Employer Retirement Savings Accounts. The ERSAs would work in essentially the same way as existing 401(k) accounts, although the proposal includes certain changes that would weaken protections for low- and moderate-income workers.
Among the most important features of the Administration’s proposal are the following:
1. It abolishes the income limits on Roth-style Retirement Savings Accounts
Under current law, taxpayers may not make deposits in Roth IRAs if their income exceeds $160,000 for married couples (and $110,000 for singles). Income limits were written into the law when Roth IRAs were established partly because economic research shows that deposits which high-income individuals make into such accounts generally do not represent new saving, but rather a shifting of saving they already have, from taxable investment vehicles to the tax-free accounts. As a result, extending IRAs to those at high income levels primarily would increase tax sheltering without doing much to increase saving.
Nevertheless, the Administration’s plan eliminates the income limits and opens up the new RSAs — which are structured identically to Roth IRAs — to people at high income levels, providing them a major new tax break.
2. It greatly increases the amounts that can be contributed to Roth-style RSAs
Today, the maximum contribution to an IRA is $3,000 for an individual plus $3,000 for a spouse, for a total of $6,000. This level is scheduled to rise to $10,000 by 2008.
These limits place a constraint only on taxpayers with high incomes or substantial wealth, since most other Americans cannot afford to put away larger amounts each year. Peter Orszag of the Brookings Institution notes that data from the IRS and other sources indicate that only about three percent to five percent of the population makes the maximum IRA contribution today. Raising the maximum amount that can be contributed to an IRA would thus directly affect only this three percent-to-five percent of the population. Simply stated, those who cannot afford even to contribute $3,000 clearly cannot afford to contribute a larger amount and thus would not benefit from an increase in the maximum contribution limit.
A Treasury Department study of IRAs, which looked at data from 1995 when the contribution limit was $2,000, concluded that "Taxpayers who do not contribute at the $2,000 maximum would be unlikely to increase their IRA contributions if the contribution limits were increased whether directly or indirectly through a backloaded [Roth] IRA.”[6]
Under the Administration’s plan, the amount that
can be contributed each year to the new Roth-style RSA would rise to $7,500
for an individual and $15,000 for a couple. A wealthy couple that cannot
currently use the Roth IRA tax break because its income is over the $160,000
income limit thus would be allowed to place at least $15,000 a year in an
RSA.
The gains to high-income families from these dramatic changes in policy — and the losses to the Treasury — would be large. For example, an affluent couple that contributed $30,000 a year and secured a seven percent rate of return would, over 20 years, be able to shelter from the income tax some $716,000 in investment income earned on these deposits.[7] Under current law, the $716,000 in income would be subject to tax. These windfall gains would place great strain on the federal budget in coming decades.
|
Economists Raise Doubts Plan Would Do Much to Increase Private Saving In an article in the February 1 New York Times, reporter Daniel Altman notes that among economists, “skepticism over the Administration’s latest proposal was widespread, even uniting academics who have long disagreed about the effects of tax-preferred plans on household savings.” Altman reports, for example, that Harvard economist David Wise and University of Wisconsin economist John Karl Scholz — two economists who often have disagreed over taxes and saving — both expressed strong doubts about the plan. Wise warned that the replacement of traditional IRAs with expanded Roth IRAs might lead to a decline in private saving. He noted that the upfront deduction offered in traditional IRAs provides an incentive for middle-income people to make deposits in those IRAs and predicted that some people would stop contributing to IRAs under the Administration’s plan because they would no longer receive an immediate tax benefit. Scholz said he thought the Administration’s proposal as a whole “would reduce private saving in the aggregate rather than increase private saving.” |
3. It establishes a new savings vehicle known as "Lifetime Savings Accounts”
In addition to the Roth-style RSAs, the proposal would create "Lifetime Savings Accounts.” These accounts would operate like Roth IRAs, except that withdrawals from the accounts could be made at any time for any purpose. These accounts, too, would be available to people no matter how high their incomes. People who could afford to do so could deposit up to $7,500 per household member into these accounts every year, in addition to deposits they made into Retirement Savings Accounts.[8]
A wealthy family of four thus could put $30,000
a year into these lifetime saving accounts and place another $15,000 a year
into Roth-style RSAs.
Moreover, the amounts placed in the LSAs and RSAs would be in addition to amounts deposited in tax-advantaged employer-sponsored retirement plans. Starting in 2006, those who can afford to do so will be allowed to deposit $15,000 a year into tax-advantaged employer-sponsored plans. Firms often make generous further contributions into these plans on behalf of business owners and highly paid executives, and federal law already permits combined employer-employee contributions into these plans of $40,000 a year for an owner or executive (or other staff members). Thus, a wealthy two-earner couple with two children would be able to benefit from contributions of as much as $125,000 a year into tax-favored accounts under the Administration’s plan — $15,000 in contributions to RSAs, $30,000 more to LSAs, and $80,000 in combined employer-employee contributions to each of the couple’s employer-based retirement accounts.
4. It shifts new IRA contributions from traditional IRAs to Roth-style RSAs
The plan would end the current situation under
which many tax filers may choose to make deposits into either traditional
IRAs or Roth IRAs. After
To understand why this constitutes a budget gimmick, it is useful to review some history. Traditional IRAs — which until a few years ago were the only type of IRA that existed — were scaled back for upper-middle- and upper-income taxpayers in the 1986 Tax Reform Act. This scaling back was part of a trade-off in the 1986 Act, in exchange for sharp reductions in income tax rates, especially for more affluent taxpayers. IRA proponents in the financial services industry and on Capitol Hill subsequently sought to weaken or undo the IRA restrictions that the 1986 law imposed, but they ran up against a seemingly insurmountable obstacle. Federal budget rules adopted in 1990 required that all tax cuts and entitlement increases be "paid for” through offsetting tax increases or entitlement cuts. Those pushing to expand IRA tax breaks could not find ways to offset the hefty costs. As a mechanism to circumvent the "pay as you go” rule, the concept of Roth IRAs was born. Because Roth IRAs would be backloaded, they would entail little cost in the years covered by the Congressional budget process. Instead, the costs would occur outside the "budget window.” Roth IRAs were designed to enable major IRA expansions to be pursued without the revenue losses having to be offset.
The Administration’s new proposal takes this type of budget maneuver and greatly enlarges it. As noted, the plan would channel all new IRA contributions into new Roth-style RSAs. Amounts that otherwise would have been contributed to traditional IRAs, with a resulting upfront cost, would be channeled into Roth-style RSAs instead, shifting the costs outside the budget window. Since the cost of any tax change that Congress considers is its cost compared to current law, shifting new IRA deposits from traditional IRAs to RSAs saves money for the government in the short term, as there will no longer be tax deductions for IRA contributions. But it results in larger revenue losses in future decades, since more money will be withdrawn tax free when people retire.
The savings this produces in the five-year budget period that the Administration’s budget covers can then be used to help offset the costs within the budget window of the other elements of the Administration’s saving package, such as the costs of eliminating income limits, increasing the amounts that may be contributed, and establishing Lifetime Savings Accounts in addition to the expanded Roth-style RSAs. The result is to make it appear as though the overall package has no cost.
In short, this is a budget gimmick that shifts the revenue losses that would have resulted from continuing to provide deductions for contributions to traditional IRAs into the future, when funds in the RSAs are withdrawn tax-free. This short-term increase in revenue is only a timing shift, however, that produces a political benefit for the Administration but increases the budget burden on future generations.
5. It provides incentives to convert traditional IRAs to Roth-style RSAs
The proposal also contains a second budget gimmick. It provides incentives for people who currently have traditional IRAs to convert those accounts now to Roth-style Retirement Savings Accounts. In converting an account, a taxpayer would withdraw the funds from his or her traditional IRA, pay tax on the amount withdrawn, and deposit the withdrawn amount into an RSA. All subsequent earnings would be tax free, as would withdrawals during retirement.
The incentives in the Administration’s plan for people to convert traditional IRAs to RSAs are designed to lead significant numbers of people to make these conversions. Those who make conversions in 2003 would be permitted to spread any taxes on the converted funds over the next four years. This results in substantially more revenues being collected during the next four years than would otherwise be the case. These added revenues can then be used to help offset the cost in the five-year budget window of the large new RSA and Lifetime Saving Account tax breaks that the proposal creates. The bottom line is that the use of these various timing gimmicks results in the package actually raising revenue over the next five years.
But the added revenues that would be collected over the next few years do not represent added revenues for the government. They are revenues that would have been collected anyway, in later years when taxpayers with traditional IRAs retired and began to withdraw the funds. In other words, this budget gimmick, like the one described above, is a timing shift: It shifts large amounts of revenue from future decades to the present. In so doing, it moves revenue from a period when the baby boom generation will be retiring in large numbers — and revenue will be desperately needed — into the next few years. This maneuver makes the Administration’s IRA/savings package look inexpensive and helps its budget seem less fiscally imprudent.
6. It weakens worker protections for employer-based pension plans
Finally, the proposal would consolidate rules for a variety of different types of employer-based retirement plans. By itself, this consolidation might constitute useful simplification. But the proposal also would repeal or weaken fundamental provisions of federal law that are designed to prevent owners and executives from concentrating an overwhelming share of a business’ retirement contributions on themselves and shortchanging their workers.
Federal pension law essentially establishes a social contract. Generous tax breaks are allowed for contributions to employer-sponsored retirement plans, including very large contributions on behalf of owners and executives. In return for these tax breaks, firms must abide by certain minimum rules designed to provide assurance that at least a modest share of a firm’s pension contributions are made on behalf of ordinary workers.
The Administration’s plan would repeal some of these protections and weaken others. The Wall Street Journal reported on February 4 that "[t]he new employer-sponsored retirement savings plan unveiled as part of the Bush budget would strip away most of the rules designed to keep employer plans from favoring highly paid employees.”[9] The likely result would be reductions in employer contributions for low- and moderate-income workers.
For example, the proposal would repeal what is known as the "top heavy” rule. This rule requires that when a small business sets up a retirement plan that directs most of its retirement contributions to top officials — and thus is a "top heavy” plan — the plan must guarantee that contributions equal to at least three percent of wages are made on behalf of low-paid workers, and that this is done without requiring those workers to come up with a matching contribution.
Separate "nondiscrimination rules” in federal law also are designed to provide adequate treatment for rank-and-file workers and to prevent egregious skewing of employer contributions to top officials and away from ordinary workers. Unlikely the "top heavy rules,” the nondiscrimination rules would not be repealed, but they would be weakened. The Wall Street Journal article quoted Harvard law professor Daniel Halperin, a leading pension expert, as stating that the result would be "less retirement savings for lower-income people.”
Perhaps most important, the proposal would likely lead over time to fewer employer-sponsored retirement plans even being offered, especially in the small-business sector. As noted above, the Administration proposal would enable owners and highly compensated executives to place huge sums in tax-sheltered retirement and saving accounts each year and to do so without having to offer a retirement plan through their firm. Currently, owners and executives must offer a plan through the workplace if they work to shelter such large sums for themselves. The Administration proposal thus breaks the social contract in this area as well.
End Notes:
[1] Treasury Department, "General Explanations of the Administration’s Fiscal Year 2004 Revenue Proposals,” February 2003.
[2]
Daniel Altman, "Accounts Chock-Full or a Plan Half Empty,” The New
York Times,
[3]
Mary Williams Walsh, "Details Given On New Plans to Aid Savings,” The
[4]
Theo Francis and Ellen Schultz, "Retirement-Savings Proposal Has Small
but Significant Changes,” The Wall Street Journal,
[5] This $3,000 limit rises to $4,000 in 2005 and to $5,000 in 2008, and is indexed to inflation thereafter. Additional "catch-up” contributions are allowed for individuals over age 50.
[6] Robert Carroll, "IRAs and the Tax Reform Act of 1997,” Office of Tax Analysis, Department of the Treasury, January 2000.
[7]Estimate
prepared by Ways and Means Democratic staff, cited in Jonathan Weisman,
"New Tax-Free Savings Plans Proposed,” The Washington Post,
[8] Contributions to Lifetime Savings Accounts are not based on earnings. Thus, parents can make contributions on behalf of their children. Contributions to Retirement Savings Accounts, like current IRAs, are linked to earnings.
[9]
Theo Francis and Ellen Schultz, "Retirement-Savings Proposal Has Small
but Significant Changes,” The Wall Street Journal,