The value of retirement benefits and tax policy

It seems that every discussion or rumor of a discussion on a “Grand Bargain” on the federal budget and entitlements includes some proposed change in taxation. Some of these proposals are modest, but most have as a basic premise that in order to raise significant revenues some fundamental changes to the tax system must be made. Any of the proposed changes currently being discussed would affect the value of current tax benefits for retirement savings.

We have published a series of articles on issues related to the budget-entitlements-comprehensive tax reform debate: Tax reform and retirement savings; Retirement savings tax incentives and the data; Tax reform and 401(k) fairness; and How efficient is the 401(k) system? In this article we consider what is at stake for participants in this debate — just how valuable are the current tax benefits to participants, and how would changes to Tax Code rules under consideration affect that value?

We will focus exclusively on defined contribution/401(k) plans where (as we have discussed in prior articles) most of the changes in rules are likely to take place. And we will focus on high margin taxpayer/participants, because the impact of changes will generally be greatest for them.

The value to participants of the current system

We’re going to begin with a base case describing the value of current retirement savings tax incentives, with which we will then compare different alternatives.

Currently a taxpayer in the highest marginal federal income tax rate bracket, 35%, can make an excludible contribution to a 401(k) plan of $17,000 in 2012 (we’re going to ignore catch-up contributions). The contribution goes into a non-taxable trust. When the contribution and any trust earnings are paid out, the taxpayer pays taxes on the entire distribution.

How much is this favorable tax treatment worth? For our base case we’re going to assume that:

The taxpayer’s marginal tax rate is the same in the year of contribution and the year of payout.

The trust earns 3% per year.

The money remains in the trust for 10 years.

In that case, given a $17,000 contribution in year 1, in year 10 the taxpayer will get a payout of $22,847, and after she pays taxes on that distribution (at a 35% rate), she will have $14,850.

If she didn’t contribute to the plan, but saved the money outside the plan, how much would she have? After paying taxes in year 1, the $17,000 is reduced to $11,050. If the taxpayer invests that amount in taxable investments paying a (before tax) return of 3%, with annual distributions to cover taxes owed on earnings, then in year 10 she will have $13,404.

Thus the difference in value between in-plan savings and non-plan savings, in year 10, is $14,850 – $13,404 = $1,446. In year 10 the taxpayer will have $1,446 more if she saves in the plan than if she saves outside the plan.

Earnings rate

These results are very sensitive to what earnings rate you pick. That’s an important fact to keep in mind in today’s environment of historically low interest rates. Here’s the same analysis, but comparing results for a 1%, 3%, and 5% earnings rate.

The results change significantly: at 1% earnings the tax savings is 1/3 of what it is at 3% earnings; at 5% earnings the tax savings is almost twice what it is at 3% earnings.

If the “new normal” is lower returns, then the value of retirement savings tax incentives is also (significantly) lower than it was in, e.g., the higher returns environment of the late 1990s.

Present value of tax benefit

So far we’ve discussed the value of saving inside a plan after 10 years. If our taxpayer saves $17,000 in a plan, then in 10 years she will have (assuming a 3% earnings rate) $1,446 more than if she saved outside the plan. What is $1,446 10 years from now worth today? That’s an important question, especially for policymakers who are trying to determine “how much we are spending (as a “tax expenditure”) on retirement savings tax incentives.”

If you discount it at 3%, $1,446 10 years from now is worth $1,076 today. So, in effect, the 2012 tax expenditure for this one high marginal tax rate taxpayer for her $17,000 contribution is $1,076.

Some nuances

That is the simple version — at least as simple as we can get it. The following factors “muddy” these results:

There are tax-favored earnings available outside a plan that cannot be captured inside a plan: capital gains tax rates (a qualified trust generally pays no taxes and all distributions are taxed at ordinary income rates); the deduction for corporate dividends; and other “tax shelters.”

There is research implying that high paid employees may forego income in order to get tax-favored qualified plan benefits, reducing the net gain shown in the chart (see our article Tax reform and 401(k) fairness).

If the taxpayer is in a lower bracket when the 401(k) benefit is distributed, that will enhance the tax benefits of in-plan savings.

And, obviously, if the participant’s tax rate is lower, that will reduce the net tax effect (in effect increasing the relative value of non-plan savings).

* * *

The foregoing analysis presents the value of current tax treatment of qualified plan retirement savings for a high marginal rate taxpayer. There are many more ways to look at the numbers — most obviously we could vary the holding period. Generally, the value of the tax benefit goes up the longer the money stays in the trust.

We would say, in sum, looking at these numbers, that the current qualified plan tax benefit is a good deal, although at today’s, lower rates of return, not really a great one. And that is a point worth keeping in mind as we move on to a consideration of the impact of changes to the current system. To repeat: as rates of return on savings and investment go down, so do the tax benefits and so does the taxpayer’s “stake” in the current system.

Effect of proposed changes in the current system — reduced marginal rates

Perhaps the most often-discussed change to the current tax system is a reduction in marginal rates paired with a “closing of tax loopholes.” Even if, in such an effort, policymakers made no change to the current tax benefits for retirement savings, a reduction in marginal rates would reduce retirement savings tax benefits.

Here’s another table, using our base case 3% earnings rate but considering two alternative “highest marginal rates:” 28% and 23%, the high and low highest marginal rate alternatives under the National Commission on Fiscal Responsibility and Reform (Simpson-Bowles Commission) “Co-Chairs’ Proposal.”

At lower tax rates, the value of the tax benefit goes down (relative to non-plan savings), but real value remains. And we would note that if as part of comprehensive tax reform other (non-qualified plan) tax benefits are reduced — e.g., if the capital gains rate is increased or the dividend deduction eliminated — then qualified plan savings become more attractive.

Effect of a cap on the value of deductions

Another proposal that has been discussed is to cap the value of qualified plan deductions. So that, even though the taxpayer/participant is paying taxes at a marginal rate of, say, 35%, the value of his deduction would be calculated using, e.g., a 28% rate. The Administration’s 2012 budget called for just such a change — a reduction in the value of itemized deductions and other tax preferences, including retirement savings tax incentives, to 28% for joint filers with income over $250,000 (at 2009 levels) and single taxpayers with income over $200,000.

A cap on the value of deductions significantly decreases the value of the tax benefit for taxpayers with marginal tax rates above the cap. For the sample employee we have been considering at a 35% tax rate, capping the deduction at 28% reduces the value of the tax benefit at the end of ten years by two-thirds, from $1,446 to $475.

Effect of a hard limit on contributions

While it is not easy to figure out exactly what sorts of caps they are talking about, there is clearly discussion amongst policymakers of reducing the dollar limits on, e.g., 401(k) contributions. The following table compares the tax benefit (currently available) with respect to a $17,000 contribution vs. a $10,000 contribution, at a 35 percent marginal tax rate.

The math here is pretty simple — a $7,000 reduction in savings, at 10 years, reduces the tax savings by around $600.

Changes in rates and contribution levels

A plausible combination is that marginal tax rates will go down and the limits on DC/401(k) contributions will be reduced. The following chart considers the value of current tax treatment – 35% highest marginal rate and $17,000 401(k) contribution limit — with a 28% highest marginal rate and a $10,000 contribution limit.

With these changes the tax benefit under the current system is nearly halved.

* * *

Let’s be clear about what we’ve been doing. We’ve been looking at one year’s tax benefit under different alternatives. The current benefit for a $17,000 contribution — at 3% interest and if it is left in the plan for 10 years — is worth $1,446 at the end of that period. Under current rules, every year that a highest marginal rate taxpayer makes that $17,000 contribution to a qualified plan (and leaves it there for 10 years), he gets the benefits we’ve discussed.

That number will be affected by several variables, critically the earnings rate and how long the contribution is left in the plan. And when we say that that is what the tax benefit is “worth,” we mean that is what it is worth relative (under different tax policy scenarios) to not contributing to the plan and instead saving in a (taxable) non-plan “vehicle.”

In our next article we are going to consider the tax effects of going to a tax credit system, as some have suggested.


The current political situation with respect to the budget, entitlements and tax reform — the possible “Grand Bargain” — is very fluid. It’s possible (but unlikely) that something may be done this year. It’s more likely that something will be done next year, possibly along the lines of one or more of the alternatives considered above.

We will continue to update you as this matter develops further.