Retirement Planning Case Study for $500K–$5M Households: A $1.5M Plan for Taxes, IRMAA, and Income

If you convert “too much” to Roth, you can accidentally spike your Medicare premiums. If you convert “too little,” Required Minimum Distributions (RMDs) can balloon later and push you (or your surviving spouse) into higher tax brackets for the rest of your life.

That’s the tension I see in real planning meetings with households who did everything right: saved diligently, built a solid portfolio, and now want a retirement paycheck that feels predictable. The problem is the paycheck doesn’t come from one place. It comes from four different account types, each with different tax rules, plus Social Security, plus Medicare rules that punish you for having income at the “wrong” time.

In this case study, I’m going to walk you through an anonymized near-retiree household with about $1.5 million spread across taxable, IRA, Roth, and cash. We’ll compare two competing retirement plans:

Plan A: Start Social Security earlier and do minimal Roth conversions.

Plan B: Delay Social Security and do planned Roth conversions while carefully managing Medicare IRMAA.

I’ll show you the tradeoffs in plain English, with simplified numbers. This is not tax advice, and your exact results will depend on your tax return, state taxes, and your specific Social Security record. But the decision logic is the point.

The household we’re planning for (anonymized, but realistic)

Meet “Mark and Dana.” Mark is 64 and Dana is 62. Mark plans to retire at 65. Dana will retire now.

They’re the kind of household I’d put squarely in the $500K–$5M “retirement complexity zone.” They’re not worried about paying the electric bill. They are worried about making a handful of irreversible choices that can quietly cost six figures over a 25–30 year retirement.

Here’s their balance sheet at the start of planning:

Taxable brokerage: $520,000 (mostly stock funds with embedded gains)

Traditional IRA/401(k): $720,000

Roth IRA: $160,000

Cash (bank + money market): $100,000

Total investable assets: $1,500,000

Spending goal:

They want $95,000 per year after tax for spending (today’s dollars), plus they’ll cover travel with “bonus” spending in good market years.

Other key facts:

Married filing jointly

No pension

They itemize some years but not reliably

They give to charity, but not enough to drive the plan by itself

They’re healthy, family longevity is “above average,” and they want to protect the surviving spouse

The planning question:

How do we fund the first 10 years of retirement (roughly ages 65–75 for Mark) while:

Keeping taxes intentionally low (not accidentally low)

Avoiding unnecessary Medicare IRMAA surcharges

Reducing future RMD risk

Keeping the portfolio risk at a level they can actually stick with

If you want the broader framework this case study sits inside, start with our pillar guide: /post/retirement-planning-500k-5m-framework.

The rules of the game (simple definitions, because the jargon matters)

Before we compare the two plans, you need four concepts. I’ll keep them simple.

1. Taxable vs IRA vs Roth vs cash

Taxable brokerage: When you sell investments, you may owe capital gains tax. Dividends can also be taxable each year.

Traditional IRA/401(k): Most withdrawals are taxed as ordinary income. Ordinary income is the same tax system that applies to wages.

Roth IRA: If rules are met, withdrawals are typically tax-free.

Cash: No market risk, but inflation risk is real. Cash is a tool, not a long-term strategy.

2. MAGI and why Medicare cares

Medicare uses a number called MAGI (Modified Adjusted Gross Income) to decide whether you pay extra premiums. Those extra premiums are called IRMAA (Income-Related Monthly Adjustment Amount).

Here’s the key: IRMAA is based on your income from two years ago. So the income you create at 63 can raise Medicare premiums at 65.

If you want the deeper explanation of what counts and how the “two-year lookback” works, see: /post/medicare-irmaa-magi-500k-5m.

3. Roth conversions

A Roth conversion is when you move money from a traditional IRA to a Roth IRA. You pay ordinary income tax on the amount converted.

You do this on purpose when your tax rate is relatively low, to reduce future taxes and future RMDs.

But conversions increase MAGI, which can trigger IRMAA. That’s the chess match.

More on the conversion/IRMAA coordination here: /post/roth-conversions-irmaa-early-retirement.

4. Sequence-of-returns risk

This is the risk that the market drops early in retirement, when you’re taking withdrawals. A bad early sequence can do more damage than a bad late sequence.

The fix is not “avoid stocks forever.” The fix is having guardrails: a portfolio you can stick with, a cash buffer, and a withdrawal plan that adapts.

We go deeper on guardrails here: /post/sequence-of-returns-risk-guardrails.

Assumptions for the case study (so you can judge the results)

I’m going to use simplified, planning-grade assumptions. Real planning uses software, tax projections, and multiple scenarios. But you don’t need a 40-page report to understand the decision.

Assumptions:

Inflation: 2.5% per year

Portfolio long-term return (nominal): 5.5% blended, with normal ups and downs

Tax law: current brackets continue (this is a planning assumption, not a prediction)

State taxes: ignored for simplicity (in real life, state tax can change the “best” answer)

Medicare: IRMAA thresholds apply; we’ll treat them as “cliffs” even though the system is tiered

Social Security: Mark’s benefit at Full Retirement Age (67) is estimated at $3,200/month; Dana’s is $2,200/month at 67

Important: Social Security numbers are examples. Your benefit depends on your earnings record.

The two competing plans (and why both sound reasonable)

Plan A: Earlier Social Security + minimal Roth conversions

The story:

Mark claims Social Security at 67.

Dana claims at 67.

They keep Roth conversions small, mostly “as needed.”

They spend from taxable and IRA as they go.

Why people like it:

It feels simple.

It avoids writing big checks to the IRS in the early years.

It reduces the chance of triggering IRMAA from conversions.

What tends to be misunderstood:

“Low taxes now” can mean “high taxes later,” especially when RMDs start.

If one spouse dies first, the survivor files as single, often at higher tax rates on the same income.

You can accidentally create a big tax problem in your 70s and 80s when your flexibility is lower.

Plan B: Delay Social Security + planned Roth conversions with IRMAA awareness

The story:

Mark delays Social Security to 70.

Dana claims at 67 (or potentially 70 depending on survivor planning).

They do intentional Roth conversions in the “gap years” between retirement and Social Security/RMDs.

They manage MAGI to control IRMAA tiers.

They use taxable and cash strategically to fund spending while converting.

Why people like it:

It can raise the guaranteed income floor later (bigger Social Security checks).

It can reduce future RMDs and future tax brackets.

It can make the surviving spouse’s tax situation less painful.

What tends to be misunderstood:

Delaying Social Security is not automatically “better.” It’s a tradeoff.

Roth conversions are not automatically “better.” They are a tax-rate arbitrage.

If you convert aggressively without watching MAGI, you can overpay Medicare premiums.

Now let’s put numbers to it.

Plan A in numbers: the “keep it simple” path

Retirement timeline:

Ages 65–67 (Mark): no Social Security yet, living off portfolio

Ages 67–70: Social Security starts at 67 for both

Age 73+: RMDs begin (current law; this can change)

Income sources in early years (65–67):

They need about $95,000 after tax.

Without Social Security, most of that comes from taxable sales and IRA withdrawals.

Here’s what typically happens in Plan A:

They sell $60,000–$80,000 from taxable (some portion is capital gains).

They pull $30,000–$50,000 from the IRA.

They keep Roth untouched “for later.”

Tax effect (simplified):

IRA withdrawals stack on top of other income and are taxed as ordinary income.

Capital gains may be taxed at a lower rate, but gains still increase MAGI.

Medicare effect:

Before Medicare starts, IRMAA isn’t in play yet.

But the income they generate at 63 and 64 can affect Medicare premiums at 65 and 66.

In this household’s case, they’re not doing big conversions, so IRMAA risk is lower.

So what’s the problem?

The problem shows up later.

By age 73, their IRA is still large because they didn’t reduce it. Even if the market is “fine,” a $720,000 IRA can easily still be $700,000–$900,000 by then depending on returns and withdrawals.

RMDs force taxable income.

Even if they don’t need the money, the IRS requires withdrawals. Those withdrawals can:

Push more Social Security into taxation

Push them into higher ordinary tax brackets

Trigger IRMAA tiers in their mid-70s

And here’s the kicker: the survivor problem.

If Mark dies first at, say, 82, Dana becomes a single filer. Single tax brackets are tighter. The same RMD + Social Security income can be taxed at higher rates.

In Plan A, the “simple” approach often creates a later-life tax squeeze.

Plan B in numbers: delaying Social Security and using the gap years on purpose

Plan B is built around a concept most retirees miss:

Your lowest-tax years are often the first years after you stop working but before Social Security and RMDs fully kick in.

Those are your “gap years.” You can either waste them (by keeping income artificially low) or use them (by converting at controlled tax rates).

Step 1: Decide what we’re trying to accomplish

For Mark and Dana, the goals are:

Increase guaranteed income later (for longevity and survivor protection)

Reduce the size of future RMDs

Keep Medicare premiums from getting needlessly inflated

Maintain a portfolio risk level they can live with

Step 2: Social Security timing choice

In Plan B:

Mark delays to 70.

Dana claims at 67.

Why this mix?

Delaying increases Mark’s benefit roughly 8% per year from 67 to 70 (plus inflation adjustments). That’s a meaningful raise on a benefit that lasts as long as he lives.

Also, the higher earner’s benefit matters most for survivor planning. When one spouse dies, the survivor keeps the larger of the two benefits.

So delaying Mark’s benefit is like buying a larger inflation-adjusted “widow(er) pension,” funded by using portfolio assets in the early years.

If you want the full breakdown of 62 vs 67 vs 70 tradeoffs, see: /post/social-security-timing-500k-5m.

Step 3: Build a Roth conversion “lane” with IRMAA awareness

Here’s the conversion strategy we tested:

Convert $60,000–$90,000 per year from Mark’s retirement date until age 70.

Then reassess once Social Security starts.

But we don’t just pick a conversion number out of thin air.

We set two ceilings:

A tax-bracket ceiling (for example, “don’t convert into a bracket we think is unattractive for this household”)

An IRMAA ceiling (for example, “try to stay under a chosen IRMAA tier unless the long-term benefit is clearly worth it”)

This is where planning becomes very household-specific.

Some households should accept an IRMAA tier for a year or two if it meaningfully reduces future RMDs.

Other households should avoid IRMAA like the plague because their conversion benefit is marginal.

The point is: IRMAA is a cost, not a moral failure. Sometimes it’s worth paying. Sometimes it’s not.

Step 4: Withdrawal sequencing to fund spending while converting

This is the part that looks like magic when it’s done well.

If Mark and Dana convert, say, $80,000 from IRA to Roth, they still need money to live.

If they also withdraw another $60,000 from the IRA to spend, their taxable income could get too high.

So instead, we fund spending primarily from:

Taxable brokerage (selling lots with attention to gains)

Cash buffer (for part of the first year)

Selective Roth withdrawals only if needed (usually later)

Meanwhile, we use IRA dollars for conversions, not spending.

That’s the sequencing concept in one sentence:

Use taxable/cash to live on while you use IRA dollars to convert at controlled tax rates.

If you want the deeper guide on which accounts to tap first and why, see: /post/withdrawal-sequencing-500k-5m.

Step 5: What happens to taxes and Medicare in the “conversion years”

Let’s make it concrete.

Assume a conversion year at age 66:

They need $95,000 after tax for spending.

They sell $85,000 from taxable (assume $25,000 of that is capital gains).

They convert $80,000 from IRA to Roth.

They have small other income.

Simplified tax picture:

The $80,000 conversion is ordinary income.

The capital gains add to MAGI.

Their total MAGI might land in a range where:

They pay a manageable federal tax bill (because they’re using a bracket on purpose)

They may or may not cross an IRMAA tier depending on the exact year’s thresholds and their deductions

This is where precision matters.

In real planning, we run the tax projection and then “dial the conversion knob” up or down to land where we want.

Step 6: What changes at 70 when Social Security starts

At 70, Mark’s Social Security is higher than it would have been at 67.

That higher benefit reduces how much they need from the portfolio every year.

And because we’ve been converting, their IRA is smaller than it would have been under Plan A.

So in their 70s:

RMDs are lower

Total taxable income is often lower or at least more controllable

IRMAA risk can be lower because you’re not forced into large IRA withdrawals

This is the “pay some tax earlier to pay less tax later” trade.

What we found: the real tradeoffs (not the sales pitch)

Here’s the honest comparison I’d give a client.

Plan A tends to win when:

Longevity is uncertain and you strongly prefer earlier guaranteed income

Your IRA is not that large relative to spending needs (so RMDs won’t be a big issue)

You’re already going to be in a low tax bracket later

You hate complexity and you will not follow a multi-year conversion plan

Plan B tends to win when:

You have a meaningful traditional IRA/401(k) balance (Mark and Dana do)

You expect at least one spouse to live into the mid/late 80s or beyond

You care about the surviving spouse’s tax situation

You want more control over lifetime taxes, not just this year’s taxes

What’s overrated

Obsessing over the single “perfect” Social Security age.

Yes, it matters. But the bigger lever for many $500K–$5M households is how you manage taxable income across decades, not just the claiming date.

What’s underrated

The gap years.

Those years are your best chance to reshape your tax future. Once RMDs and large Social Security benefits are in place, your flexibility shrinks.

What’s commonly misapplied

Roth conversions done without a Medicare plan.

I’ve seen households do a big conversion at 63 or 64, feel proud, and then get hit with higher Medicare premiums at 65 because of the two-year lookback. That’s not “wrong.” It’s just uncoordinated.

Portfolio risk: how the investment mix changes the odds of success

Now let’s talk about the part everyone wants to skip until the market drops.

Mark and Dana’s initial portfolio was roughly 70% stock / 30% bonds, plus $100,000 cash.

They told me they were “comfortable with risk.” Most people say that in a calm market.

The real question is different:

Can you stick with your plan if the market drops 20% in the first two years of retirement?

Because if you can’t, the plan doesn’t matter. You’ll abandon it at the worst time.

In this case study, we tested two risk postures:

Risk posture 1: Keep 70/30

Risk posture 2: Shift to 60/40 and formalize a cash buffer

Here’s what changed with posture 2:

They held roughly 18–24 months of planned withdrawals in cash and short-term bonds (not forever, just as a buffer).

They kept enough stock exposure to fight inflation over a 30-year retirement.

They created a rule: in a major market drawdown, they temporarily reduce taxable sales and use the buffer instead.

This matters because Plan B (delaying Social Security) requires you to spend more from the portfolio in the early years.

That’s not bad. It’s just a fact.

So if you delay Social Security, you need a sequence-of-returns plan. Otherwise, you’re taking more withdrawals exactly when the market might be down.

That’s why I pair Social Security delay conversations with guardrails. If you want to see how we think about those guardrails, start here: /post/sequence-of-returns-risk-guardrails.

Medicare IRMAA: how we kept conversions from turning into a premium penalty

IRMAA is one of the most frustrating retirement “gotchas” because it feels like a stealth tax.

Here’s the plain-English version:

If your income is above certain thresholds, Medicare adds a surcharge to your monthly premiums.

Two important details:

It’s based on MAGI from two years ago.

It’s tiered. Crossing into a higher tier can raise premiums for the whole year.

In Mark and Dana’s plan, we did three practical things:

1. We mapped the two-year lookback

If Medicare starts at 65, then ages 63 and 64 tax returns matter.

So we avoided “surprise income” in those years when possible.

2. We separated “conversion years” from “Medicare-sensitive years” when it made sense

Sometimes the best conversion window is before Medicare.

Sometimes it’s after Medicare starts but under a chosen tier.

The right answer depends on your income sources and how big your IRA is.

3. We treated IRMAA as a line item, not a boogeyman

If a conversion saves $25,000 in future taxes but costs $2,500 in extra Medicare premiums for one year, that may be a good trade.

But if it saves $3,000 and costs $2,500, that’s not a win.

This is why we run the numbers.

If you want the detailed explanation of what triggers IRMAA and what counts in MAGI, see: /post/medicare-irmaa-magi-500k-5m.

The retiree questions that matter most right now (answered directly)

How can I optimize my retirement withdrawals to minimize taxes?

Think in decades, not years.

A common mistake is minimizing this year’s taxes by avoiding IRA withdrawals and conversions. That can backfire when RMDs arrive.

A better approach is:

Use taxable and cash strategically in early retirement

Use planned IRA withdrawals/conversions to “fill” a reasonable tax bracket each year

Protect Roth for later-life flexibility (and for the surviving spouse)

The mechanics are covered in detail here: /post/withdrawal-sequencing-500k-5m.

When should I start taking Social Security to maximize benefits?

“Maximize benefits” depends on what you mean.

If you mean the biggest monthly check, delaying increases the check.

If you mean the best lifetime outcome, it depends on longevity, portfolio size, and whether you’re protecting a spouse.

In Mark and Dana’s case, delaying Mark to 70 improved the survivor benefit and reduced pressure on the portfolio later.

But if your health is poor or you have a smaller portfolio, claiming earlier can be reasonable.

We break down the tradeoffs here: /post/social-security-timing-500k-5m.

How can I avoid higher Medicare premiums due to IRMAA?

You can’t “avoid” IRMAA in the abstract. You can only manage the income that triggers it.

The practical levers are:

Control the size and timing of Roth conversions

Be careful with large one-time income events (big IRA withdrawals, large capital gains, property sales)

Coordinate income two years before Medicare starts

More detail here: /post/roth-conversions-irmaa-early-retirement and /post/medicare-irmaa-magi-500k-5m.

What is the best order to withdraw funds from my taxable, IRA, Roth, and cash accounts?

There isn’t one universal order.

But for many $500K–$5M households, a common pattern is:

Use cash for short-term needs and as a volatility buffer

Use taxable for spending in early retirement (while managing capital gains)

Use IRA strategically (for bracket-filling withdrawals and/or Roth conversions)

Preserve Roth for later years, big expenses, and survivor flexibility

The “why” matters more than the order. Full guide: /post/withdrawal-sequencing-500k-5m.

How can I balance portfolio risk to ensure a stable retirement income?

Stability comes from a combination of:

A risk level you can stick with

A cash/short-term bond buffer so you’re not forced to sell stocks in a downturn

A withdrawal plan that can flex (even slightly) when markets are rough

If your plan assumes you’ll behave perfectly in a bear market, it’s not a plan. It’s a hope.

Practical takeaways from this $1.5M case study (what I’d want you to copy)

1. Treat the first 5–10 years as a distinct planning phase

This is where Social Security timing, Medicare start dates, and Roth conversions collide.

2. Don’t confuse “tax-free” with “best”

Roth is powerful, but converting is not free. You’re choosing when to pay taxes.

3. Watch the two-year Medicare lookback

If you’re 63–64 and Medicare starts at 65, your income decisions right now can change your premiums soon.

4. Make Social Security a household decision, not an individual decision

For married couples, the higher earner’s timing is often the anchor because of survivor benefits.

5. Pair any Social Security delay strategy with sequence-of-returns guardrails

Delaying can be smart. Delaying without a risk plan is fragile.

If you want the full coordinated approach that ties these pieces together, revisit the pillar: /post/retirement-planning-500k-5m-framework.

Your action list: build your own Retirement Tax & Income Map

If you’re within about five years of retirement (or newly retired) and you have a mix of taxable, IRA, Roth, and cash, here’s the action list I’d use before making big moves.

1. Inventory your accounts by tax type

List balances in taxable, IRA/401(k), Roth, and cash.

2. Estimate your “income floor” at 70

What will Social Security be at 70 for each spouse?

Any pension?

This tells you how much the portfolio must produce later.

3. Identify your gap years

Years between retirement and when Social Security/RMDs meaningfully increase income.

Those are your planning years.

4. Choose a conversion ceiling

Pick a target tax bracket you’re comfortable filling.

Then check whether that level of income risks pushing you into an IRMAA tier.

5. Build a withdrawal sequence that supports the conversion plan

If you’re converting from IRA, consider funding spending from taxable/cash in those years.

6. Stress-test the first two retirement years

Ask: what if the market drops 20% early?

Do we have a buffer?

Do we have a rule for where withdrawals come from?

7. Put Medicare on the same page as taxes

Write down when Medicare starts and which tax years feed the IRMAA lookback.

8. Re-check annually

This is not a “set it and forget it” plan. Tax brackets, thresholds, and markets change.

If you want help turning your account mix into a coordinated plan, I’ll invite you to request what I call a Retirement Tax & Income Map: Social Security timing, Medicare/IRMAA awareness, Roth conversion lanes, withdrawal sequencing, and portfolio risk guardrails—built around your numbers.

Book an appointment here: http://grapewealthmanagement.salesmate.io/meetings/#/grapewealthmanagement/user/1

AI/automation disclosure: This article was created with the assistance of AI tools and reviewed for accuracy, clarity, and usefulness by Grape Wealth Management before publication.