How to Build a Tax-Efficient Withdrawal Order for Early Retirement and Market Downturns

The wrong withdrawal order can cost real money. A retiree who pulls from a traditional IRA too early may trigger avoidable ordinary income tax; one who drains taxable assets too aggressively may lose flexibility when markets fall. The cleanest-looking plan on paper is often the ugliest plan in a bear market.

The hard part is that the “best” sequence changes with the year. In 2026, a married couple filing jointly can remain in the 0% federal long-term capital-gains band through $98,900 of taxable income, but a few thousand dollars of extra income could also reduce Affordable Care Act subsidies or push more income into the 15% capital gains bracket ^[2][3]. That is why withdrawal planning is not just a tax problem. It is a cash-flow problem, a market-risk problem, and a calendar problem all at once.

The first dollar should usually come from the account that buys you the most flexibility

Most investors start with a rule of thumb: taxable first, then tax-deferred, then Roth. That is not crazy. It is just incomplete. The real question is which account gives you the most optionality after taxes, penalties, and benefit cliffs are counted. A taxable brokerage account can fund spending with basis, low-gain lots, or cash. A traditional IRA can be efficient later, but every dollar withdrawn is ordinary income. A Roth IRA is tax-free on qualified withdrawals, but it is also the scarcest asset in the set because future tax-free growth is hard to replace ^[1].

For early retirees, flexibility often beats purity. If you need to bridge the years before age 59½, taxable assets and Roth contributions are usually the least constrained sources. Roth conversion ladders can work, but they require a five-year waiting period for converted principal and careful coordination with income limits and penalties ^[1]. The catch is simple: the account with the lowest current tax cost is not always the account with the highest strategic value.

That is why a withdrawal order should be built around marginal tax rates, not account labels. A $40,000 withdrawal from a traditional IRA may be cheap in a low-income year. The same withdrawal can be expensive if it collides with Social Security, Medicare premiums, or ACA subsidy calculations later. The sequence matters because the tax code is full of cliffs, not smooth slopes.

For a broader framework on where to hold assets before you retire, see how to build a tax-efficient asset location plan across taxable, IRA, and Roth accounts. Withdrawal order and asset location are cousins. They should not be designed separately.

Source: IRS Publication 590-B and IRS retirement plan guidance ^[1].

Three failure modes of a naive withdrawal order

A naive plan usually fails in one of three ways. First, it ignores the tax bracket you are trying to fill. Second, it ignores the market regime you are in. Third, it ignores future forced income, especially RMDs. Those failures interact. A withdrawal that looks efficient in January can become a tax trap by December if the market drops and you need to sell appreciated assets at the wrong time.

Failure mode one is bracket blindness. Long-term capital gains and qualified dividends are taxed differently from ordinary income, and the 0%, 15%, and 20% federal long-term capital gains brackets are tied to taxable income thresholds that change with inflation ^[2]. If you have room in the 0% bracket, realizing gains deliberately can be rational. If you do not, selling appreciated taxable assets may be worse than taking ordinary income from a traditional IRA in a low-income year. That sounds backward to many investors. It is not.

Failure mode two is sequence-of-returns risk. A retiree who sells stocks after a 25% drawdown locks in a permanent loss of capital. The same withdrawal rate can be survivable in a flat market and destructive in a bad one. That is why withdrawal order should be paired with a cash reserve policy and a rebalancing policy. If you need a refresher on the mechanics of drawdowns, see drawdowns: why they matter more than returns and how to build a portfolio stress test that actually changes decisions.

Failure mode three is future forced income. RMDs start at age 73 for many retirees under current law, and the IRS Uniform Lifetime Table determines the annual percentage ^[1]. If you let traditional balances grow untouched for too long, you may end up with large mandatory withdrawals later, exactly when you would prefer more control. That is the uncomfortable implication: sometimes the tax-efficient move today is to pay some tax now to avoid a larger tax bill later.

A simple decision tree for taxable, traditional, Roth, and cash reserves

Here is the sequence I would use as a starting point for many early retirees. It is not universal. It is a decision tree, not a law.

1. Spend cash reserves first if the market is down and your reserve is sized for at least 6 to 12 months of spending. Cash is expensive in real terms, but it is cheap insurance against forced selling. For the tradeoff between cash and short-duration bonds, see when to use bonds, when to use cash, and when a short-duration ETF beats both.
2. Use taxable basis and low-gain lots next if you can fund spending without realizing large gains. Taxable accounts let you choose lots, which is a real advantage. Investors who ignore lot selection leave money on the table.
3. Harvest gains deliberately when you have bracket room. If your taxable income is low enough to stay in the 0% long-term capital gains band, realizing gains can reset basis at little or no federal tax cost ^[2].
4. Tap traditional accounts in low-income years, especially before Social Security and before RMDs begin. This is often the best window for partial Roth conversions ^[1].
5. Use Roth last for ordinary spending, unless doing so prevents a worse tax event or preserves ACA subsidies. Roth is the cleanest money. Clean is not always cheapest.

This sequence changes when the market is stressed. In a severe drawdown, the first goal is not tax minimization. It is avoiding forced equity sales. That is why a withdrawal policy should specify a “stress mode” and a “normal mode.” If you want a framework for making rules-based decisions instead of improvising, the logic is similar to systematic vs. discretionary investing: pre-commit to the rule before emotions show up.

Source: AIBROKER analysis based on IRS tax rules and ACA income mechanics ^[1][3].

ACA subsidies can make a low-tax year surprisingly expensive

Early retirees often focus on federal income tax and miss the Affordable Care Act. That is a mistake. Marketplace premium tax credits are based on household income relative to the federal poverty level, and the subsidy formula can change sharply as income rises ^[3]. A few thousand dollars of extra ordinary income or realized gains can reduce subsidies enough to wipe out the tax savings from a clever withdrawal sequence.

The same problem appears with capital gains. A retiree who thinks, “I am in the 0% capital gains bracket, so I should sell more stock,” may accidentally push modified adjusted gross income high enough to reduce ACA support. The tax code does not care that the gain was “only” 0% federally. The subsidy formula still sees income.

That is why withdrawal planning for early retirement should be done on a full-year basis, not month by month. If you are under 65 and buying insurance on the exchange, the right question is not “Which account is cheapest?” It is “Which withdrawal keeps my total household income inside the best band?” That band may be narrower than you think.

There is a second trap. Some retirees realize gains in December to use up the 0% capital gains bracket, then discover that the extra income also affects state taxes, Medicare later, or the taxation of other benefits. The federal bracket is only one layer. State rules can be harsher, and they vary widely. A withdrawal plan that ignores state tax is half a plan.

For investors who want to understand how tax-aware rebalancing works in taxable accounts, see how to build a portfolio rebalancing policy that minimizes taxes, costs, and behavioral mistakes. The same lot-selection logic applies to withdrawals.

Roth conversions are a tool, not a religion

Roth conversions are often sold as a universal good. They are not. They are a trade: you pay ordinary income tax now in exchange for tax-free withdrawals later. That trade is attractive when your current tax rate is low and your future rate is likely higher. It is less attractive when you are already near a subsidy cliff or when you need the converted money before the five-year clock runs out ^[1].

The best conversion window is usually the gap between retirement and the start of Social Security and RMDs. That window can be surprisingly valuable because ordinary income may be low enough to fill the 10%, 12%, or 22% brackets without spilling into higher rates. But the conversion should be sized to the bracket, not to the account balance. Converting too much is a classic self-inflicted wound.

Here is the judgment: most investors overconvert. They see a large traditional balance and assume they should “get it into Roth” as fast as possible. That is often wrong. A measured conversion schedule that respects tax brackets, ACA income, and cash needs is usually better than a heroic one-time conversion. The goal is not to eliminate all future taxes. The goal is to control them.

If you want to compare this with the broader retirement glide path, the logic connects to the glide path decision. Asset allocation and withdrawal sequencing are two sides of the same retirement balance sheet.

Worked example: Suppose a retiree needs $60,000 of spending. They have $200,000 cash and taxable basis, $700,000 in a traditional IRA, and $300,000 in Roth. In a normal year, they might take $20,000 from taxable basis, $25,000 from the traditional IRA to fill a low bracket, and leave Roth untouched. In a bear market, they might take the full $60,000 from cash and taxable basis, then delay IRA withdrawals until markets recover. The tax bill is not identical across those years, but the sequence protects the portfolio from forced selling.

A withdrawal policy should specify what happens in a 20% drawdown

Retirement plans fail in bad markets because they were written for average markets. Average markets are not the problem. The problem is the year when stocks fall 20% to 30% and spending still has to happen. That is when a withdrawal policy needs a stress clause.

A good stress clause says three things. First, how much cash or short-duration bond reserve you will hold. Second, how long you will avoid selling appreciated equities after a drawdown. Third, what tax bracket or income band you will target while markets are weak. Without those rules, retirees improvise. Improvisation is expensive.

One useful approach is to define a “red zone” drawdown threshold. If the portfolio is down more than 15% from its recent high, withdrawals come from cash and low-volatility assets first. If the drawdown exceeds 25%, you may temporarily reduce spending or pause Roth conversions. That is not a prediction. It is a rule. The point is to avoid selling the wrong asset at the wrong time.

This is where risk measurement matters. A retiree who only watches average return is blind to the path. If you want a deeper framework for thinking about path risk, see risk measurement and Sharpe vs. Calmar. Calmar-style thinking is closer to retirement reality because drawdowns matter more than smooth averages.

The hidden tradeoff is that a strict stress rule can leave tax savings on the table in calm years. That is fine. A retirement plan should be robust first and elegant second.

A one-page implementation checklist beats a perfect spreadsheet

Retirement withdrawal planning gets lost in complexity because the variables multiply fast. You do not need a 40-tab model to start. You need a one-page checklist that forces the right questions in the right order.

1. Estimate annual spending and separate essential from discretionary costs.
2. List account balances by type: taxable basis, taxable gains, traditional, Roth, and cash.
3. Estimate this year’s taxable income before withdrawals.
4. Check the 0% and 15% long-term capital gains bands and the ACA income range.
5. Decide whether this is a normal year or a stress year.
6. Set a target ordinary-income bracket for traditional withdrawals or Roth conversions.
7. Choose which lots to sell in taxable, if any, and document the reason.
8. Review RMD exposure for future years and adjust conversion pace.

That checklist is enough for most households. If you want a more formal process for documenting rules, the structure is similar to writing an investment policy statement you will actually follow. The document should fit on one page if you expect to use it.

One more judgment: the best withdrawal plan is usually the one you can execute in a bad year without second-guessing yourself. A brilliant plan that collapses under stress is worse than a decent plan you can follow.

Checklist: Before each year starts, confirm your spending target, taxable income estimate, ACA status, RMD exposure, and the account you will tap first if markets fall 15%.

Withdrawal sequencing must account for required minimum distributions under current IRS rules rather than relying on a fixed order copied from another retiree. ^[6]