A few days ago, in a MarketWatch opinion piece, Alicia H. Munnell addressed a question posed by the headline, “Should we rethink how we tax Social Security benefits?” My answer is, as it has been since 1983, when the inclusion of some social security benefits were first included in gross income, a resounding “Yes.” Of course, the more important question is, “How?”
Munnell suggests that the model for taxing social security benefits should be how 401(k) plans are taxed. Yet she also concludes that no more than 50 percent of social security benefits should be taxed. She bases this conclusion on two facts. First, the portion contributed by the employer into the social security trust fund is not included in the employee’s gross income. Second, the portion contributed by the employee, which equals the portion contributed by the employer, is included in the employee’s gross income.
Munnell, however, misses a third, very important fact. She includes in her analysis on the difference between the taxation of traditional 401(k) plans and the taxation of Roth 401(k) plans. In the traditional plan, the employee is not taxed on employee contributions to the plan, and is taxed on the amounts withdrawn during retirement. In the Roth plan, the employee pays tax on what is contributed to the plan and does not pay tax on the amounts withdrawn during retirement. Here’s the catch. Even though the contributions to these plans earn income from whatever investments the plan makes, because the plans are tax-exempt, no taxes are paid while the plans are earning income on the accumulated contributions. In the traditional plan, that investment income is taxed when it is included in the amounts that are withdrawn. In the Roth plan, that investment income never gets taxed. To me, that is a flaw. Munnell also claims that if the tax rate does not change between the working years and the retirement years that the tax treatment of the traditional Is “equivalent.” However, not only does this claim ignore the non-taxation of the investment income in a Roth plan, it also ignores the time value of money, because deferring the tax payment until retirement is itself a benefit to the employee.
Determining the appropriate amount of Social Security benefits that should be taxed must begin with the definition of gross income. Simply put, gross income is the amount of a person’s income. Income is the person’s increase in economic wealth that has been clearly realized, and it is included in gross income unless it qualifies for a specific exclusion from gross income. Putting aside the niceties of what “clearly realized” means, a concept that students in a basic tax class struggle to comprehend and that requires deep, intensive reading and analysis of more than a few cases, and that in the context of social security is more a matter of timing than anything else, to the extent a social security recipient receives more than what the recipient contributed that was already taxed, the recipient has gross income.
It is easy to determine the extent to which a social security recipient has gross income. The question is timing. And that question is easily answered. Suppose a social security recipient contributes $50,000 to social security over a working lifetime. Basic tax law principles dictate that the recipient should not be taxed on the first $50,000 received in social security benefits. Thereafter, all of the payments should be taxed. It’s that simple. The key, of course, is knowing how much the person has contributed. The Social Security Administration has a record of that information. So why wasn’t that approach adopted? The official explanation was that the information was not available or, at best, not easily retrieved by the Social Security Administration. Whether or not that was true 37 years ago – I don’t think it was – it certainly isn’t the situation today. The real reason in 1983 was the need for revenue, to offset the adverse consequences of the 1981 tax cuts, and the approach I advocated would delay revenue from the taxation of social security benefits for several years as people simply recovered what they had paid in. Granted, it would have been a bit more complicated than that, to deal with the fact that some people already receiving social security benefits would have already received what they contributed and would be immediately subject to tax. But the simple approach I championed was rejected in favor of what was a more complicated gyration of computations made even worse a few years later.
There is another flaw in the current system. A person who dies before receiving benefits equal to what was contributed, and who has sufficient modified adjusted gross income to be taxed on social security benefits, can end up being taxed even though they lost money by contributing more than they got back. And offsetting that flaw is yet another, which is that someone who lives long enough to receive more than what was contributed ends up not being taxed on what unquestionably is a clearly realized increase in economic wealth.
So when Munnell suggests that fifty percent of social security benefits “might be viewed as the appropriate share of benefits to include in adjusted gross income,” she fails to address any of the flaws in how social security benefits are taxed. Under her approach, a person dying shortly after retirement would continue to be taxed on amounts that are not income, and a person living long enough would continue to receive income free of tax that should be taxed. I should note at this point that the concern about taxing social security recipients who have low income should be, and can be, addressed not through the “base amount” and “adjusted base amount” nonsense of current law but simply by providing a standard deduction high enough to protect low-income individuals from taxation no matter the source of the income.
Several of the comments to Munnell’s opinion piece remind me of how much ignorance about social security runs rampant. One person claimed, “Most of SS is just a return of your own money, therefor tax free makes the most sense.” That is absolutely untrue, except for the unfortunate folks who die shortly after beginning to receive social security benefits. Another person made a similar claim, stating, “Social security should be treated like a Roth IRA. not taxable at all you paid income tax on that money already.” No, you did not, again, unless you unfortunately die soon after retiring. Fortunately, other persons commenting on the article pointed out these misconceptions, suggesting that people can calculate what they, or a retiree they know, has received in social security benefits and compare that to what the person contributed. Many of the comments focused on other social security issues, not the taxation question, and then addressed other tax issues, both sensibly and with demonstrated ignorance, and though I could write several book chapters separating the comments reflecting good understanding of economics, finance, and tax from those reflecting something other than a good understanding, at the moment I will leave that for others.
As a practical matter, any attempt to make changes to the taxation of social security benefits will open the door to the continuing attempt to privatize social security and put its control into the hands of private equity funds, oligarchs, and wealthy financiers. Some of the comments to Munnell’s opinion piece reflected the unrealistic expectation that no one loses money making private investments. How quickly the world has forgotten Bernie Madoff, the folks at Enron, the wizards at Adelphia, to say nothing of the investment advisors with good intentions but inadequate skill sets. Americans have become so eager to purchase the Brooklyn Bridge. It is sad what happens when the money addicts meet those ignorant of economics, finances, and taxes. Very sad.