In October my paycheck was about fifteen hundred dollars bigger than September's, and I could not tell you why.

Nothing had changed. No raise, no bonus, no adjustment to my 401(k), no change to my benefits. Gross pay identical. Federal withholding identical. Virginia withholding identical. I went line by line down two statements looking for the thing that moved, and when I found it, it wasn't an addition. It was a subtraction that had stopped.

Social Security tax. Present in September, gone in October. Not reduced. Gone.

I had hit the wage cap, which I did not know existed.

What Actually Happened

Social Security is funded by a payroll tax of 6.2% on your wages, matched by another 6.2% from your employer. But only up to a point. SSA calls it the contribution and benefit base, and for 2026 that ceiling is $184,500. Earn a dollar above it and that dollar is not subject to the tax at all.

So the tax runs until you've paid the maximum, which in 2026 is $11,439, and then it switches off for the rest of the year. In January it switches back on against a new ceiling, which for 2027 rises to $192,300.

Medicare works differently and the contrast is the whole essay in miniature. Medicare's 1.45% applies to every dollar you earn with no ceiling whatsoever, and above $200,000 an additional 0.9% kicks in. So in the same month that my Social Security tax vanished, a new Medicare surcharge appeared.

One tax has a roof. The other doesn't, and actually gets steeper.

The net effect was more take-home pay, which I should be honest about enjoying for roughly a day before it started bothering me.

I Already Wrote This Essay and I Got It Wrong

Last year I wrote that the American tax code punishes labor and rewards ownership. I stand behind every word of it. Wages face a top marginal rate of 37% plus payroll tax while long-term capital gains top out at 20% and the effective rate for the very wealthy often lands in single digits. The code is a moral document and it says effort is a cost and ownership is a virtue.

But I priced payroll tax in that essay as a flat 7.65% riding on top of wages. I put it in a table that way. And that is wrong.

It isn't flat. It stops.

Read that last column again. The rate falls as income rises. Not the dollar amount, the rate. A teacher earning sixty thousand dollars pays six times the rate on her wages that someone earning a million pays on theirs.

That is a regressive tax. Not regressive in the loose rhetorical sense people use when they mean "unfair," but in the technical sense: the percentage declines as the base grows. And it is the second largest tax most working Americans pay.

I wrote an entire essay arguing that the code is structured to protect people who already have money, and I missed the cleanest available piece of evidence. Not because it's hidden. It's published on a government website. I missed it because I had never earned enough to see it happen, and nobody had ever explained it to me.

Which is its own kind of answer about how this works.

Me, at Both Ends of It

Here's the part that actually got under my skin.

I spent most of my service as an E-3 at Robins Air Force Base. Base pay was somewhere around $2,400 a month per the Department of Defense pay table, call it roughly twenty-nine thousand a year, plus a housing allowance and a subsistence allowance on top. In 2022, the Social Security wage cap was $147,000.

I was never going to touch that ceiling. Not in that job, not in ten years of that job. Every single dollar of taxable pay I earned in uniform was taxed at the full 6.2%, for every month of every year, with no stopping point, because the stopping point was built for someone else's income.

Same man. Same tax. Same country. Four years later the tax clocks out in October and the last two months of the year arrive untouched.

I want to be scrupulous about one wrinkle, because it complicates the comparison in a way that's worth knowing rather than hiding. Military housing and subsistence allowances aren't taxable income, so they weren't subject to payroll tax either. My taxable military wage was essentially just base pay. That's a genuine tax advantage of military compensation and I'm not going to pretend otherwise.

It doesn't rescue the structure. The point isn't that I had it worse than I remember. The point is that the tax had no ceiling in my life until my income got large enough for a ceiling to be relevant, which is precisely backwards from how we talk about every other tax in this country.

What Social Security Actually Is

I should admit that when I started pulling on this I wasn't certain what Social Security even was. I had a vague sense it was a retirement thing, probably running out, possibly for poor people. I was funding it at the maximum rate and could not have explained it. I suspect that's extremely common, and it's the same thing I was getting at when I wrote that you can only dream of what you've been shown.

So, plainly.

It's social insurance, not a savings account. Your payroll taxes aren't deposited anywhere with your name on it. They pay today's beneficiaries, the way today's workers will pay yours. In 2025, by SSA's own accounting, about 185 million people paid in and about 70 million people received benefits.

You qualify by working. Generally forty credits, which is roughly ten years of covered employment, not necessarily consecutive. Your benefit is calculated from your highest thirty-five years of earnings and the age you claim. You can start at 62 with a permanently reduced monthly check, full retirement age is 67 if you were born in 1960 or later, and waiting until 70 increases it.

It is not means-tested. High earners collect it. You are not paying into a program you're excluded from.

And Medicare is not Medicaid, which is the confusion I was carrying. Medicare is health coverage you become eligible for at 65 regardless of wealth, and it isn't free; Part B carries a monthly premium. Medicaid is the program with financial eligibility rules. Two different things, and conflating them is how people end up believing their payroll taxes fund somebody else's welfare rather than their own insurance.

Administrative overhead, incidentally, was 0.4% of expenditures last year. Whatever else is wrong with this program, nobody is losing the money to bureaucracy.

"It Won't Exist" Is Wrong, and the Truth Is Worse in a Different Way

Everyone my age has absorbed the idea that Social Security won't be there. That's not what the numbers say, and the actual numbers deserve to be quoted precisely.

The 2026 Trustees Report projects that the retirement and survivors trust fund depletes its reserves in the fourth quarter of 2032, at which point incoming revenue would cover 78% of scheduled benefits. The combined funds last until 2034, at 83%.

Depletion doesn't mean the tax stops arriving. It means the reserve buffer is gone and the program pays out of current receipts, which covers most but not all of what was promised. The honest sentence is "a 22% benefit cut in 2032 absent legislation," not "it will vanish."

But don't take too much comfort from that, because the trend is going the wrong way. The seventy-five-year actuarial deficit is now projected at 4.42% of taxable payroll, up from 3.82% in the previous report. Reserves fell $160 billion in a single year. Costs have exceeded non-interest income since 2010.

So this is a real problem on a visible clock, with a known arrival date, and a cap that switches off for high earners every autumn.

The Rationale, and Why It Doesn't Hold

The cap isn't an oversight. It has a design logic and the argument deserves to be stated properly before being disagreed with.

Social Security was built as contributory insurance. Your contributions are tied to your earnings, and so is your benefit. Earnings above the cap don't count toward your benefit either, so the cap isn't purely a gift; it's a boundary on both sides of the ledger. And the benefit formula is strongly progressive, replacing a much larger share of a low earner's wages than a high earner's. The redistribution happens on the payout side rather than the collection side.

That's a coherent system. I understand why it was designed that way in 1935.

It still doesn't hold up, for a reason the design argument never addresses. The cap's symmetry assumes the benefit is the point of the contribution. For someone earning well into six figures, it isn't. I am not going to be relying on a Social Security check. The program is not insurance for me in any practical sense, and the reason my taxes should keep running isn't that I'm buying something. It's that I can afford to fund a floor under people who are going to need one, and the floor is 22% short with six years on the clock.

Framed that way the symmetry argument collapses, because it answers a question nobody is really asking. Nobody defends the mortgage interest deduction by noting that renters don't get it either.

I'll say the thing directly. A tax that shuts off in October for the highest earners and runs every month of the year for everyone else is indefensible. The actuarial tidiness of it is not a defense. It's an explanation, and people routinely mistake one for the other.

My Fix, and Where It Broke

My first instinct was to flip it. Exempt the first chunk of income entirely, let the people earning the least keep their 6.2%, and run the tax on the high end where the cap currently lets it expire.

Then I tried to actually specify it, and two things went wrong.

The cliff. My first version was: under $150,000 you pay nothing, over $250,000 you pay 6% on everything. Which means the person earning $249,999 pays nothing and the person earning $250,001 pays about fifteen thousand dollars. One dollar of extra income triggers a five-figure tax bill. That's not a policy, it's a trap, and it would distort behavior around the threshold immediately.

The fix is marginal brackets, the same way income tax already works. Nothing on the first $150,000, something on the next hundred thousand, the full rate only on dollars above $250,000. Then no single additional dollar ever costs more than the rate on that dollar.

The arithmetic. This is the one that actually humbled me. Removing the cap raises revenue. Exempting the first $150,000 of everyone's wages loses revenue. Those are two separate reforms pointing in opposite directions, and I'd been treating them as one idea because they felt like the same moral impulse.

And the loss is enormous, because the overwhelming majority of American wages sit below $150,000. You cannot replace the payroll contributions of most of the workforce with a surcharge on a small number of high earners by asserting that high earners have more. That's a quantity question, not a fairness question, and it needs the total dollars of wages in each band rather than a count of people.

I don't know whether my brackets balance. I'd need the actual wage distribution to say, and I'm not going to pretend an intuition is a model. What I do know is that removing the cap alone, without exempting anyone, is a serious proposal that SSA's actuaries score as closing a large share of the shortfall. That part is real and it's on the table.

So I'll hold the position and drop the plan. The cap should go. What replaces it on the low end is a question I can raise and can't yet answer, and I'd rather say that than publish a number I made up.

The Part Worth Remembering

There's no villain in this one, which is what makes it stick.

The cap was a design choice made by people solving a 1935 problem with 1935 assumptions, and it has been quietly indexed upward every year since without anyone being asked whether the shape is still right. It's in the law. It's on a public website. It generated a fifteen-hundred-dollar swing in my own paycheck and I had to go hunting to find out why.

That's the thing I keep running into from every direction. The most consequential rules in a system are almost never the controversial ones. They're the ones that have been there so long that nobody experiences them as decisions anymore.

A tax that gets easier the more you earn is a decision.

Somebody made it. It can be made differently.