Cyrus Reed: Hey, good to be back — how was the rest of your week, before we get into something that's been quietly stressing me out?
Iris Holm: Fine. What's stressing you out.
Cyrus Reed: The Roth versus Traditional question. Because someone asked me — just, a friend, over dinner — and I started to answer and then I stopped, because I realized I was about to give advice that assumes I know what tax rates will look like in thirty years. And I don't. Nobody does. And that assumption is doing all the work.
Iris Holm: The assumption is the entire product.
Cyrus Reed: So — okay, let me start with a person. Priya, thirty-two years old, eleven p.m., 401(k) enrollment form on her laptop, deadline at midnight. Traditional or Roth. Just those two boxes.
Cyrus Reed: So no jargon — here's what each box actually means. Traditional: the money goes in before taxes hit it, you get the tax break now, and then when you retire and pull it out, that's when the IRS collects. Roth: you pay taxes on the money tonight, it grows, and when you take it out decades from now, you owe nothing. Permanently tax-free.
Iris Holm: And the present-value equivalence proof says those two are identical — if her rate never changes.
Cyrus Reed: Exactly that. The math is — wait, it's almost unsettling how clean it is. Same rate in, same rate out, same after-tax wealth. The government's share of the account is the same either way. Which means Priya isn't actually making an investment decision. She's making a bet on her own future tax bracket.
Iris Holm: So the only thing that matters is whether her tax rate goes up or down.
Cyrus Reed: That's — yeah. That's the entire question. Everything else is noise.
Iris Holm: And the framing — 'which account is better?' — that's not a performance question. It was always a tax-timing question in disguise.
Cyrus Reed: But that clean version — the 'same rate in, same rate out' version — it just got a lot more complicated, because Congress passed the Tax Cuts and Jobs Act in 2017 and overnight repriced the whole calculation for millions of people.
Iris Holm: That's the live example. TCJA cut the top marginal rate to 37%. If you were a high earner in 2018, suddenly Roth contributions looked mathematically attractive in a way they hadn't before. Not because your situation changed. Because Congress changed the law.
Cyrus Reed: Wait — so acting on that window is itself a bet that Congress won't reverse it?
Iris Holm: You're trading one uncertainty for another. And the statute sunsets in 2026. So the 'obvious' Roth move today becomes the expensive move if rates revert.
Cyrus Reed: That's — okay, that's unsettling. Because I've heard people say 'go Roth now, rates are low' as if it's just obvious, but they're actually making a political forecast dressed up as a financial plan.
Iris Holm: Legislative roulette. And it's not theoretical — look at the record. U.S. federal income tax rates since 1913: under 10% to over 90%. That's not noise. That's the documented behavior of the system Priya is betting against.
Cyrus Reed: Over ninety percent. What year?
Iris Holm: 1944. Top marginal rate hit 94%. Not a rounding error.
Cyrus Reed: Ninety-four. That's — wait, someone who opened a retirement account in, I don't know, 1935, assuming low rates, and then — yeah. The math just detonated on them.
Iris Holm: Which is exactly why the Roth IRA exists. The Taxpayer Relief Act in 1997 created it. Congress built a separate account structure precisely because future-rate uncertainty is real enough to deserve its own vehicle. That's not a financial advisor's opinion. That's Congress acknowledging the risk in statute.
Cyrus Reed: So the Roth IRA is — wait, actually, I want to sit with that for a second — it's Congress hedging against its own unpredictability? Like, they built the tool because they know they can't be trusted to keep rates stable?
Iris Holm: Frankly — yes. And Priya's advisor telling her 'go Traditional, you'll be in a lower bracket at retirement' isn't financial planning. It's a political forecast with a 30-year time horizon and no error bars.
Cyrus Reed: And that advisor probably believes it — that's the thing that keeps snagging me. Because the 'lower bracket in retirement' story isn't wrong in the abstract, it's wrong specifically for the people who are most likely to hear it and act on it.
Iris Holm: Name the two failure modes.
Cyrus Reed: Wait, actually, start with the 27-year-old software engineer. She's in a low bracket right now, her advisor says 'go Traditional, you'll earn less in retirement,' and that's — that's almost certainly backwards. Her income is going up. She's at the bottom of her career curve. Roth is structurally the better fit and someone just talked her into the wrong box.
Iris Holm: Income trajectory risk cuts both directions.
Cyrus Reed: Right — but then flip it. The 55-year-old partner at a law firm, she goes Traditional because 'I'll drop brackets when I stop billing hours.' And that feels safe, it feels conservative. Except she's been maxing contributions for twenty years and the balance is — I mean, it's large. And then the IRS shows up.
Iris Holm: Required Minimum Distributions.
Cyrus Reed: She doesn't need the money. She has a pension, Social Security, maybe some rental income. But the IRS mandates withdrawals from her Traditional IRA starting at 73 regardless. She didn't choose to take income that year. She's forced to.
Iris Holm: So the person who was most disciplined — biggest balance — is most exposed to the trap?
Cyrus Reed: That's exactly it. The RMDs mechanically generate taxable income she never planned for. And suddenly the bracket she was going to drop into — she's above it. The discipline built the trap.
Iris Holm: Roth IRA has no RMDs. Ever. The assets stay invested, tax-free, indefinitely.
Cyrus Reed: No way — none at all?
Iris Holm: Zero. And the Roth 401(k) extends that same logic into the employer plan — same contribution limits as a traditional 401(k), but Roth after-tax treatment. Career disruptions, inheritance, a spouse's income changing — thirty years of forecast error compounds. The Roth 401(k) is how you hedge that inside a workplace plan.
Cyrus Reed: Which — and I want to come back to this — means the decision you make at 55, when you actually know your balance and your income sources, probably matters more than the box Priya checked at midnight. There's a version of this where you can course-correct, and that's the part we should get into.
Iris Holm: The assumption that you locked it in at 25 is wrong. That's the reframe.
Cyrus Reed: But wait — course-correct HOW? Because the box Priya checked at midnight, that's not permanent, but it kind of feels permanent. Like, once the money is in a Traditional IRA, it just... sits there accruing this tax liability she can't see.
Iris Holm: Roth conversions. You move Traditional IRA balances into a Roth. Pay the tax in the year you convert. Done.
Cyrus Reed: Okay but — when does that actually make sense to do? Because you're voluntarily paying tax early, which sounds like the thing we said Traditional lets you avoid.
Iris Holm: Low-income years. Gap year. Early retirement before Social Security kicks in. You convert when your marginal rate is temporarily lower than it'll be when RMDs force withdrawals. That's the window.
Cyrus Reed: So someone retires at 62, no RMDs yet, Social Security hasn't started — that's actually a tax trough. And they could be converting Traditional balances into a Roth for eleven years before the IRS forces anything.
Iris Holm: That's the decision that matters more than the one at 25. Now — here's where I want to push. Splitting contributions across both a Traditional IRA and a Roth IRA from the start builds that optionality in. You don't need to predict. You just don't go all-in on one forecast.
Cyrus Reed: And Fidelity — I mean, they're running this for millions of people. They can operationalize 'hedge both.' Why is the default framing still 'pick one'?
Iris Holm: 'Pick one' is a simpler product story. But look — someone going all-in on Traditional isn't being conservative. They're making an active bet that their income falls and Congress doesn't raise rates. That's not a defensive posture. That's two simultaneous political and personal income forecasts.
Cyrus Reed: Wait — that reframe is — huh. Because 'I'll be in a lower bracket' sounds humble, right? Sounds cautious. But you're saying it's actually — no, it's an optimistic bet on your own declining earnings.
Iris Holm: And high earners can't even access the Roth IRA directly — the phase-out exists. So the backdoor Roth strategy comes in, which carries its own policy risk if Congress closes it. The architecture constrains the choice before the personal math applies.
Cyrus Reed: So the real sophistication isn't the choice at 25 — it's the conversion strategy at 55?
Iris Holm: The decision you deferred at 25 becomes the one that matters most at 55. Structure flexibility in early. Convert during the trough. The marginal tax rate comparison is still the central variable — you just get to pick when you make it.
Cyrus Reed: I keep thinking about Priya. It's past midnight now. She's been sitting with this for — I don't know, an hour? And she hasn't solved the tax-rate forecasting problem, because nobody can solve it. But she understands what the problem actually is. And that's... I mean, that's actually different from where she started.
Iris Holm: The form lets her split. She checks both boxes.
Cyrus Reed: Yeah. Some Traditional, some Roth. Not because she solved anything — because she decided not to go all-in on a single 30-year forecast about her own income and whatever Congress does next. She just... closes the laptop.
Iris Holm: The interesting question was never which account wins. It was how much of the decision she gets to keep making over time.
Cyrus Reed: That's — yeah. That's it.
Iris Holm: Roth conversions exist. Low-bracket years exist. The initial box isn't permanent.
Cyrus Reed: She doesn't know that tonight. But — wait, actually, she doesn't need to. That's the part that makes the quiet version of this feel okay. She bought optionality without knowing the word for it.
Iris Holm: Good enough for midnight.
