Redistributive Policies Hidden in the US Retirement System
Note to readers: I am more than a bit worn down by writing about the ongoing political insanity in the US (see, for example, my columns from this past Thursday and Friday), so today's column reflects my conscious decision to take a brief respite by disappearing into policy wonkitude. And what could be wonkier than fiscal policy and retirement in the United States? Join me, won't you?
In an article in the Osgoode Hall Law Journal in late 2024, I noted that "total payments to beneficiaries of those three components for the month of December 2023 were: 97.8 billion USD for old-age, 8.8 billion USD for survivors, and 11.9 billion USD for disabled persons, covering 52.7 million, 5.8 million, and 8.5 million people respectively." The two smaller parts of Social Security are thus not in fact "small," but they are dwarfed by what most people think of when they hear the words social security. That article, as noted above, constitutes my one foray into writing about any other aspect of Social Security
In a recent Dorf on Law column, "Are Emerging Political Threats to Social Security Enough to Change Retirement Strategies?" I noted that in the old-age benefits system, eligible workers are able to choose when to begin receiving their retirement benefits, no earlier than their 62nd birthday and no later than their 70th, with a "full retirement age" (FRA) not exactly halfway between the two (67 for everyone born from 1960 onward, and lower FRA's for earlier birth years as part of a phased-in increase of the FRA from 65 to 67 years). The FRA is a pivot point for some changes in benefits calculations, but it is not otherwise meaningful. That is, the only thing that matters for the worker is the knowledge of what they will receive in benefits thereafter (adjusted for inflation, which is a key feature of the system) compared to what they could have received by choosing a different starting date to collect benefits.
The empirical finding that I described in that column is that the conventional wisdom is correct that people who can afford to delay receiving benefits (and whose life expectancies are not uniquely short) should -- as a matter of pure dollars and sense (pun intended) -- wait until age 70. Even though starting to receive benefits earlier than that last possible date does provide money that can be invested in the meantime, the rate of return on those earlier years of saved benefits would have to be extraordinarily/unrealistically high in order to make early receipt a winning strategy.
Even so, I noted that my personal decision -- years before my 70th birthday -- was to begin receiving benefits. Why? Because the chaos of the Trump regime makes it unclear whether the system will continue to exist at all. Side note (as I also described in that earlier column): If the system is left alone, the worst that can happen is that benefits starting in about 2035 will be roughly 20 percent lower than the formulae would otherwise call for, but the system would be fully sustainable at that somewhat reduced level. No, Social Security is still -- no matter what Republican doomsayers have been spouting for decades -- not doomed.
My choice to start collecting my benefits early was therefore a (relatively low-stakes) bet on the possibility that the Republicans will soon achieve their long-cherished dream of destroying the best social insurance program ever devised. This is a bet that I want to lose, but for the first time in its 91 years of existence, Social Security's political future no longer appears to be guaranteed.
The primary reason that I am probably making a losing bet is that Trump's base is heavily dependent on Social Security (and Medicare, which I discuss below), and they -- completely correctly -- do not think of their entitlement program benefits as a dreaded "welfare" payment. On the other hand, Trump seems not to care that his base is being harmed by his economic policies more generally, just as he is impressively casual about having suddenly become a proponent of wars of choice and regime change. Anything is possible.
Let us imagine, however, that Social Security and Medicare survive the Trump regime relatively intact. My purpose in the balance of this column is to focus on an aspect of the Medicare program that interacts with Social Security in a surprisingly progressive way -- where "progressive" here means redistributing resources in a Robin Hood sense. As I conceded above, I have written almost nothing about Medicare to date (minor exceptions here and here), just as I have mostly ignored the second two parts of Social Security itself. The hidden gem that I will describe here is a method by which the Medicare system complements and enhances the progressivity of the Social Security system.
Some Medicare basics will be helpful to ground the analysis (ignoring many details): Part A provides coverage for hospital expenses with no premiums or co-pays, starting at age 65. Part B provides coverage for doctor visits, but there is a monthly premium that beneficiaries must pay to the government on a monthly basis. (Part B coverage is limited to 80 percent of total costs, so people can buy so-called MediGap or Supplemental private insurance.) Part C is private "Medicare Advantage" plans that are a rip-off and bad policy to boot (as John Oliver explained this past October) that substitutes for traditional Medicare coverages. Part D is the program for prescription drugs, again provided by private insurers that charge a monthly premium.
Part B is the key here. The monthly premium payments are generally paid by having the Social Security system reduce a person's monthly retirement benefit check. (If a person is signed up for Medicare Part B but not yet receiving benefits, which does happen, then of course there is no monthly benefit to tap.) In fact, much of the administrative work for Medicare is handled by Social Security staff, to the point where people who need assistance with a wide range of Medicare questions are instructed to call Social Security's toll-free service number.
This means that a person who is receiving retirement benefits and is enrolled in Part B sees only a monthly check that is slightly smaller than it would have been if they were not enrolled in Part B. What does "slightly" mean in the previous sentence? I have worked out a reality-based scenario for a hypothetical Citizen X, where costs are based in part on location, sex (males are more expensive than females), lifetime incomes, when the citizen signed up for the two programs, and so on. My Citizen X would be receiving about $3900 per month in Social Security old-age benefits while paying about $250 per month in Part B premiums, thus receiving a net monthly check of $3650 per month.
But then comes IRMAA. Who is she? The Income-Related Monthly Adjustment Amount sets up a little-known progressive add-on to the Social Security and Medicare universe. The government publishes IRMAA's structure (see page 2 of this document for 2024), and the important point is that Part B premiums can go up -- way up -- for people with high incomes. I decided to make Citizen X a high earner, and the numbers that I assumed resulted in an additional $475 per month IRMAA on top of the $250 that those with lower incomes pay, meaning that he would pay almost three times as much for Part B, and his monthly check would therefore be $3175 rather than $3650.
Moreover, the Social Security benefit structure itself is already refreshingly progressive. Although Social Security's payroll taxes are notoriously regressive (charged at a flat rate up to $184,500 in earned income in 2026, with zero additional taxes paid by anyone with salaries above that cap), the payout structure is very progressive. That is one of the reasons that anti-Social Security conservatives claim that the system is "unfair," because it redistributes from their target audience (rich people) to everyone else.
IRMAA, however, increases the impact significantly. And again, it is administratively seamless, because the monthly benefit check is net of Part B payments, and Part B payments include IRMAA.
Why is that a big deal? Two of the evergreen neoliberal policy proposals to "save" Social Security are to eliminate the tax cap and to tax benefits. I have argued at various times that the former idea is not terrible but hits a relatively well-off group without truly taxing "the rich" (who mostly do not earn money in ways that are subject to Social Security taxes at all). The latter idea, taxing retirement benefits, is politically tricky for a number of reasons, starting with the odd optics of the government giving benefits and then taking some of them back. It also makes the system seem less universal, at least in some people's minds.
But if we correctly view both Social Security and Medicare as a unified retirement system, IRMAA is a way to "tax" high-income retirees without seeming to tax them at all. Because the public debate around all of these questions is often about form and not substance, having what amounts to a progressive tax integrated into the medical benefits program is rather interesting.
I am planning to write more about this issue at law review length soon, exploring other ways in which the country's retirement system could be made more progressive. Today's column, then, is my first go at explaining the quirks of the current system and how they can work together in ways that are excellent as a matter of policy but are not obvious.
- Neil H. Buchanan