What If You Just Paid the 10% Early-Withdrawal Penalty?

A post that circulated widely in August 2026 made an argument you have probably heard in some form: 401(k)s and Roth IRAs are a scam, because who wants to wait until they are 60, old and tired to touch their own money.

The impulse behind it is reasonable. Restricted money is worth less than unrestricted money, and someone at 30 who expects to change careers, buy a house, or stop working at 45 has real uses for capital long before 59½. The usual response is to list the ways around the restriction: the Rule of 55, 72(t) payments, a Roth conversion ladder. Those all work, and we have written the mechanics of each.

This guide asks a different question. Suppose you ignore every one of them. You retire at 45, you claim no exception, you liquidate a traditional 401(k) in one go and you hand the IRS the full 10% additional tax on the whole balance. How far behind does that leave you against someone who skipped the account entirely and used a taxable brokerage instead?

On plausible assumptions, it does not leave you behind at all.

Two corrections before the math

The threshold is 59½, not 60. The IRS defines early distributions as those “you receive from a qualified retirement plan or deferred annuity contract before reaching age 59½.”¹ Six months is not the point, but if the argument rests on a specific age it is worth getting the age right.

A Roth IRA is a strange thing to pick as an example of locked-up money. Under the ordering rules in Publication 590-B, your regular contributions come out first, and the IRS states plainly: “You don’t include in your gross income qualified distributions or distributions that are a return of your regular contributions from your Roth IRA(s).”² The money you put in is reachable at any age, for any reason, with no tax and no penalty. Restrictions attach to earnings and to converted amounts, each of which carries its own five-year clock.

Neither correction is the interesting part. Set them aside and grant the argument its strongest form: assume you will need the money early, you will not do any planning, and you will simply eat the penalty.

The question nobody asks

Comparing accounts has to start from pre-tax income, because that is where the fork actually is. $10,000 of salary buys $10,000 inside a traditional 401(k). The same $10,000 buys about $7,600 in a taxable brokerage account, because you pay income tax on it first. Compare $10,000 to $10,000 and you have given the taxable account a head start it never had.

With that fixed, here is the crossover year at which a traditional 401(k) liquidated early, penalty paid in full, overtakes a taxable brokerage account. Assumptions: 7% total return, a 3.25% dividend yield (the current 30-day SEC yield on SCHD, a widely held dividend ETF),³ 15% on qualified dividends and long-term gains, no state tax.

Read the first row again. Someone who contributes in the 24% bracket while working and withdraws in the 12% bracket after retiring early is better off in the traditional account from the very first year, having paid the penalty in full. There is no waiting period to endure. The deduction they collected on the way in is larger than the tax and penalty they pay on the way out.

This is a measurement of the worst case, and it should be read that way. Nobody is suggesting you plan to liquidate a retirement account and pay a penalty. The point is narrower and harder to argue with: if even the worst available exit beats never contributing, then the decision to contribute survives the liquidity objection, and what remains to work out is how much to keep in a taxable account alongside it.

Why the bracket drop decides it

The intuition most people have, and the one the original argument gestures at, is that the tax drag of a taxable account is what makes retirement accounts worth the lockup. For a dividend-heavy portfolio that drag is real: dividends are taxed the year they are paid whether you wanted the cash or immediately reinvested it, which is exactly why a dividend strategy makes tax-advantaged space more valuable rather than less.

But it is not what moves the answer. Hold the tax-rate path fixed at 24% today and 22% at withdrawal, and vary only the yield:

Four years of difference across the entire plausible range of yields. Now hold the yield fixed and change the tax-rate path instead, and the break-even swings from year 18 to year 1. The dividend drag is a real cost that compounds for decades, and it is second order here. What determines whether the deferral pays is whether your marginal rate falls between the year you contribute and the year you withdraw.

That has a practical consequence. Someone planning to retire at 45 is describing a future with very low taxable income: no salary, spending funded from savings, and years of headroom in the 10% and 12% brackets before Social Security or required distributions begin. That is the exact profile the first row of the table describes, contributing at 24% and withdrawing at 12%.

What the penalty costs once you amortize it

A 10% penalty sounds enormous because it is quoted as a percentage of the balance. Amortize it over the holding period and it becomes a number you can compare with anything else. On the assumptions above, the 10% additional tax is equivalent to an annual drag of:

• 138 basis points a year if you hold for 10 years
• 69 basis points a year over 20 years
• 46 basis points a year over 30 years

Set that against what plans charge. The BrightScope and ICI study of 59,012 plans holding $5.5 trillion put total plan cost at 0.85% for the average plan and 0.52% for the average participant, with plans under $1 million in assets averaging 1.29% and plans over $1 billion averaging 0.27%.⁴ Over a long horizon, the one-time 10% penalty is a smaller drag than the ongoing fees of a below-average 401(k) plan. If you are going to be angry about a number, the fees are the better target, and unlike the penalty you pay them whether or not you ever touch the money early.

The Department of Labor’s standing illustration makes the same point about fees from the other direction: $25,000 growing at 7% for 35 years reaches $227,000 if fees take 0.5% a year and only $163,000 if they take 1.5%, a 28% reduction from one percentage point.⁵

Independent confirmation

Stephen Horan solved the same break-even problem in the Financial Services Review in 2004 and reached the same place. His finding for a saver whose bracket falls, stated verbatim:

“For investors in the 25% tax bracket dropping to the 15% tax bracket, a traditional IRA with an early withdrawal penalty is a superior investment vehicle for any time horizon.”⁶

Horan also concluded, more broadly, that “a 10% early withdrawal penalty is not substantial enough to discourage investors from using IRAs for non-retirement purposes.”⁶ His least favorable case runs the other way: with a flat 25% bracket and a taxable portfolio holding a single fund whose entire gain is deferred until a terminal sale, his break-even stretches to 28.5 years. That is the modern broad-market ETF held forever and never sold, which is a real strategy and the strongest version of the taxable case.

When the taxable account genuinely wins

The result is not universal, and a guide that pretended otherwise would be worth less than one that draws the boundary.

• Your bracket does not fall. If you expect the same or a higher marginal rate at withdrawal, the deferral buys you little and the penalty is a straight loss. At a flat 22% the break-even is 18 years, and above about 24% at withdrawal it can stop arriving at all.
• You need the money within about five years. Short horizons give compounding no time to outrun the penalty. Money for a house, a business, or a sabbatical inside five years belongs somewhere liquid.
• Your plan is expensive. A 1.29% plan is giving back most of the tax advantage every year. That is an argument for contributing to the match and then routing the rest elsewhere, and for reading the rollover decision guide when you leave.
• You are already in the 0% capital-gains bracket. For 2026 that runs to $49,450 of taxable income for a single filer and $98,900 for a couple filing jointly.⁷ If your qualified dividends and long-term gains are taxed at zero, the taxable account has no drag to escape.
• You have no liquid assets at all. Tax efficiency is a second-order concern for someone who would have to sell at a bad moment or borrow at 24% to handle a broken transmission.

Why you would still avoid the penalty if you can

Everything above measures the floor. If you are going to reach retirement money early, there are several ways to do it without the 10%, and they are strictly better than paying it:

• Separation from service at 55 or later. The IRS exception applies when “the employee separates from service during or after the year the employee reaches age 55.” It covers qualified plans. The IRS exceptions table marks the IRA column no, which is a good reason not to roll an old 401(k) into an IRA reflexively.⁸
• Governmental 457(b) plans have no early-withdrawal penalty at all. The IRS states that distributions from a governmental 457(b) “are not subject to the 10% additional tax except for distributions attributable to rollovers from another type of plan or IRA.”⁸ If you have access to one, the entire premise of the argument disappears.
• Substantially equal periodic payments under 72(t). Notice 2022-6 permits any rate “not more than the greater of (i) 5% or (ii) 120% of the federal mid-term rate.”⁹ The schedule is rigid and busting it is expensive, so this is a commitment rather than a convenience.
• Roth contributions, at any time. Per the ordering rules above.

There is one asymmetry the calculator surfaces that is easy to get backwards. If you liquidate an entire Roth balance early, the earnings are taxed as ordinary income and penalized, on money you already paid income tax on once. On the assumptions used here, over 30 years that leaves the Roth behind both the traditional account and the taxable brokerage. Take contributions first and leave the earnings where they are.

Who cashes out, and when people retire

Two pieces of evidence cut against reading any of this as encouragement.

First, people who tap retirement accounts early mostly are not running an optimization. Vanguard’s How America Saves 2025 found that among participants who left a job in 2024, 29% cashed out entirely. That headline needs its denominator stated, because the same report shows those cash-outs were only 5% of assets: 94% of the dollars stayed invested. Small balances dominate the headcount. About 80% of people with balances under $1,000 cashed out, against 2% of those with $250,000 or more.¹⁰ The people paying the penalty are largely not the people for whom this arithmetic was run.

Second, the premise that you will control the timing is shakier than it sounds. In the 2026 EBRI and Greenwald Retirement Confidence Survey, workers gave a median expected retirement age of 65 while retirees reported a median actual age of 62, and 46% of retirees left earlier than planned. Of those, 41% cited a hardship such as a health problem or disability.¹¹ The error runs in both directions: 12% of workers expect to retire before 60, while 29% of retirees did. Planning around a confident date at either end is planning around a number that moves.

Does any of this increase saving?

Worth separating two questions that get conflated. Whether tax-favored accounts raise national saving is contested. Whether an individual who was going to save anyway should use them is not.

On the contested one, the Journal of Economic Perspectives ran both sides back to back in the same 1996 issue. Poterba, Venti and Wise concluded that “contributions to these accounts represent new saving in large part.”¹² Engen, Gale and Scholz concluded, thirty pages later, that “little if any of the contributions to existing saving incentives have raised saving.”¹³ The journal staged it as an unresolved disagreement and it has not fully resolved since.

The strongest modern evidence comes from Denmark, where Chetty and coauthors used 41 million observations and found that “each $1 of government expenditure on subsidies increases total saving by only 1 cent,” because roughly 85% of people are passive savers who never respond, while the 15% who do respond mostly shift money from taxable accounts rather than saving more.¹⁴ That result is regularly quoted as evidence against retirement accounts, which inverts it. It is a finding about the government’s return on subsidy spending, not about whether an individual benefits. The same paper found that automatic employer contributions raise total saving by about 80 øre per krone, with effects persisting more than ten years. The active savers who shift assets into tax-advantaged accounts are, in the paper’s own framing, the wealthier and more financially sophisticated ones. They are doing the thing this guide describes.

On the behavioral side, Madrian and Shea’s study of a single firm switching to automatic enrollment remains the cleanest demonstration that plan design beats plan economics: tenure-matched 401(k) participation went from 37% to 86%, and 61% of the automatically enrolled cohort sat at every single default, against 1% of every other cohort.¹⁵

One last piece of theory, with a caveat attached. Beshears and coauthors modeled how much liquidity a savings system should offer when people are present-biased, and found that some illiquidity improves welfare rather than harming it. Their 2020 working paper described a three-account optimum including one with roughly a 10% early-withdrawal penalty, and noted the resemblance to the American system. That specific result did not survive revision: the version published in the Journal of Financial Economics in 2025 describes a simpler two-account optimum and drops the comparison.¹⁶ The general claim survived: some illiquidity raises welfare in their model, present-biased households included.

Frequently asked questions

Is it really true that I can withdraw Roth IRA contributions anytime?

Your regular contributions, yes, at any age and for any reason, free of tax and penalty. Converted amounts each carry their own five-year clock, and earnings are only tax-free in a qualified distribution. The ordering rules in Publication 590-B take contributions out first, which is what makes this work in practice.

Does this mean I should plan to pay the 10% penalty?

No. It means the penalty is a smaller threat than it looks, so it should not drive the decision about whether to contribute. If you expect to need the money before 59½, use the Rule of 55, a 72(t) schedule, a governmental 457(b), Roth contributions, or a taxable account sized for the bridge. Paying the penalty is the fallback that makes the downside survivable.

What about the 10% on a Roth?

Contributions come out free. Earnings withdrawn early and non-qualified are taxed as ordinary income and hit with the 10%. That combination is why liquidating an entire Roth balance early is the worst of the three options modeled here, despite the Roth being the best of the three when held to a qualified distribution.

How much can I contribute in 2026?

$24,500 of elective deferrals to a 401(k), plus an $8,000 catch-up at 50 or older, or $11,250 for ages 60 through 63. IRAs take $7,500, plus a $1,100 catch-up. Total annual additions to a defined contribution plan cap at $72,000.¹⁷

Does a state income tax change the answer?

It usually strengthens it, because a state deduction today against no state tax later is the same bracket-drop logic applied twice. Several states exempt retirement income entirely. The calculator applies one flat state rate to both ordinary income and capital gains, which is a simplification.

What if my 401(k) has terrible funds?

Contribute enough to capture the full employer match, since that is compensation rather than an investment decision, then weigh the plan cost against the tax benefit for dollars beyond it. A plan charging 1.29% a year is consuming most of the advantage this guide describes.

Key takeaways

• Compare from pre-tax income. $10,000 of salary is $10,000 in a traditional 401(k) and about $7,600 in a taxable account. Skipping that step is how the comparison gets rigged.
• With a bracket drop, the penalty never costs you the lead. Contributing at 24% and withdrawing at 12%, the traditional account wins from year one even after paying the 10% in full.
• The bracket path decides it, and the dividend drag is secondary. Varying the yield across its whole plausible range moves the break-even four years. Varying the tax path moves it seventeen.
• Amortized, the penalty is 46 basis points a year over 30 years. The average 401(k) plan charges 0.85%, and small plans charge 1.29%, which is the larger of the two numbers.
• Retiring at 45 means years with almost no taxable income. That puts you in the 12% bracket at withdrawal, which is the row where the deduction pays off fastest.
• Do not empty a Roth in one transaction. Contributions come out free; earnings come out taxed and penalized.

Related guides

• How to Withdraw From a 401(k) or IRA Before 59½ Without the Penalty: the companion piece, and what to read if you would rather not pay it.
• Roth vs. Traditional 401(k) and IRA: the break-even retirement tax rate, which is the variable this guide shows to be decisive.
• Roth Conversion Ladder: the planned version of early access, with the five-year clock mapped out.
• Why I Avoid SCHD: what a forced dividend yield costs in a taxable account, with a 20-year drag simulator.
• You Have One Household Portfolio, Not One Per Account: which account should hold which asset once you are using all of them.
• 403(b) vs. 457(b) vs. 401(k): the governmental 457(b) and its missing early-withdrawal penalty.
• Don’t Automatically Roll Over Your Old 401(k): why rolling to an IRA can cost you the Rule of 55.

Sources and method

1. IRS, Topic no. 558, Additional tax on early distributions.
2. IRS, Publication 590-B, chapter 2, “Ordering Rules for Distributions.”
3. Schwab Asset Management, Schwab U.S. Dividend Equity ETF (SCHD). 30-day SEC yield 3.25% as of August 5, 2026; expense ratio 0.060%. Schwab updates this daily.
4. BrightScope and Investment Company Institute, A Close Look at 401(k) Plans, 2022, published March 2025. 59,012 plans, $5.5 trillion in assets.
5. U.S. Department of Labor, Employee Benefits Security Administration, “A Look at 401(k) Plan Fees,” September 2019.
6. Stephen M. Horan, “Breakeven holding periods for tax advantaged savings accounts with early withdrawal penalties”, Financial Services Review, vol. 13, no. 3 (2004), pp. 233–247.
7. IRS, Revenue Procedure 2025-32, maximum capital gains rate amounts for 2026.
8. IRS, Retirement topics: Exceptions to tax on early distributions. Source of the separation-from-service row, the IRA column, and the governmental 457(b) footnote.
9. IRS, Notice 2022-6, which modified and superseded Rev. Rul. 2002-62.
10. Vanguard, How America Saves 2025, figures 115–116, data year 2024.
11. EBRI and Greenwald Research, 2026 Retirement Confidence Survey, Fact Sheet 2. Fielded January 2026, n = 2,544.
12. James M. Poterba, Steven F. Venti and David A. Wise, “How Retirement Saving Programs Increase Saving,” Journal of Economic Perspectives, vol. 10, no. 4 (1996), pp. 91–112.
13. Eric M. Engen, William G. Gale and John Karl Scholz, “The Illusory Effects of Saving Incentives on Saving,” Journal of Economic Perspectives, vol. 10, no. 4 (1996), pp. 113–138. Published in the same issue as the preceding entry.
14. Raj Chetty, John N. Friedman, Søren Leth-Petersen, Torben Heien Nielsen and Tore Olsen, “Active vs. Passive Decisions and Crowd-Out in Retirement Savings Accounts: Evidence from Denmark,” Quarterly Journal of Economics, vol. 129, no. 3 (2014), pp. 1141–1219.
15. Brigitte C. Madrian and Dennis F. Shea, “The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior,” Quarterly Journal of Economics, vol. 116, no. 4 (2001), pp. 1149–1187. Single firm, roughly two years of data, so treat the magnitudes as illustrative.
16. John Beshears, James J. Choi, Christopher Clayton, Christopher Harris, David Laibson and Brigitte C. Madrian, “Optimal illiquidity,” Journal of Financial Economics, vol. 165 (2025), article 103996. The three-account result with a 10% penalty appears in the earlier NBER Working Paper 27459 (2020) and not in the published version.
17. IRS, IR-2025-111 and Notice 2025-67, 2026 retirement plan limitations.

Method. All break-even figures come from the calculator above, which runs in your browser. One dollar of pre-tax income is followed through three wrappers: a taxable account funded after income tax, with dividends taxed annually and reinvested at basis and the accumulated gain taxed on liquidation; a traditional account compounding untaxed and distributed as ordinary income plus the 10%; and a Roth funded after income tax, with earnings taxed and penalized only in the non-qualified case. The model uses a single flat marginal rate rather than a bracket stack, applies one state rate to both ordinary income and capital gains, and excludes the net investment income tax, fund turnover, plan fees and any step-up in basis at death. Those simplifications mostly favor the taxable account, so the break-evens shown are conservative.

Editor’s note

Educational content, not tax or investment advice, and not a recommendation to withdraw from any retirement account. The social media argument that prompted this is paraphrased rather than attributed, because the point is the reasoning and not the person. Citations verified against primary sources on August 7, 2026.