Sign In

Skip Navigation LinksOneFPA > Journal > Using Roth Conversions to Add Value to Higher-Income Retirees’ Financial Portfolios

by William Reichenstein, Ph.D., CFA; and William Meyer

Executive Summary

  • This study shows how financial advisers can add value to higher-income households by recommending Roth conversion strategies before required minimum distributions (RMDs) begin and before tax rates are scheduled to increase in 2026.
  • In general, higher-income households (for the purposes of this paper) have incomes that are too high to avoid paying taxes on less than 85 percent of Social Security benefits if they follow the conventional wisdom withdrawal strategy, but their incomes are high enough that they should consider how the withdrawal strategy from their financial portfolio will affect the size of their Medicare premiums.
  • This study demonstrates how the taxation of Social Security benefits causes taxpayers to pay marginal tax rates within a wide income range that is 150 percent or 185 percent of their tax brackets. It then explains that increases in income-based Medicare premiums can cause spikes in higher-income households’ marginal tax rates.
  • Two cases demonstrate the steps higher-income households can take to lower the size of their lifetime income taxes and the
    size of their lifetime Medicare premiums. Ultimately, higher-income households should consider making Roth conversions in their first few retirement years, before RMDS begin, and before tax rates are scheduled to increase in 2026.

William Reichenstein, Ph.D., CFA, is head of research at Social Security Solu­tions Inc. and Retiree Inc. He is professor emeritus at Baylor University and has published more than 190 articles and written several books including Social Security Strategies (third edition) with William Meyer and Income Strategies.

William Meyer is CEO of Retiree Inc. in Overland Park, Kansas. He and William Reichenstein developed software to combine a smart Social Security claiming decision with a tax-efficient withdrawal strategy (incomesolver.com).

JOIN THE DISCUSSION: Discuss this article with fellow FPA Members through FPA's Knowledge Circles​​​​. ​​​

Financial advisers provide their clients important support, knowledge, encouragement, and guidance to achieve their financial goals. Research shows that advisers can also provide considerable value to clients’ bottom line—their portfolios.

A 2013 paper by Morningstar researchers (Blanchett and Kaplan 2013) introduced the concept of “gamma” to quantify how smarter financial planning decisions can add value to clients’ portfolios. And 2016 Vanguard research (Kinniry et al. 2016) suggested that using relationship-oriented services (what Vanguard calls “advisor’s alpha”) can add about 3 percent to clients’ net returns.

Reichenstein and Meyer (2018) explained how financial advisers can add value specifically to middle-income households’ accounts by: (1) adjusting when they begin their Social Security benefits; and (2) recommending a tax-efficient withdraw strategy from their financial portfolio.

This study explains how financial advisers can add considerable value to higher-income households by recommending Roth conversion strategies before RMDs begin and before tax rates are scheduled to increase in 2026 (“middle-income” and “higher-income” households are defined later).

The next two sections explain how the taxation of Social Security benefits and income-based Medicare premiums can sharply increase retirees’ marginal tax rates, where marginal tax rate denotes the taxes paid on the next dollar of income. This study then presents two cases that demonstrate that higher-income households can reduce both their lifetime income taxes and their lifetime Medicare premiums by making Roth conversions before RMDs begin and before tax rates are scheduled to rise in 2026, as legislated in the Tax Cuts and Jobs Act (TCJA) of 2017.

Taxation of Social Security Benefits

Reichenstein and Meyer (2018) explained how Social Security benefits are taxed. The taxable portion of Social Security benefits depends on a household’s Provisional Income (PI). In formula, PI = Modified Adjusted Gross Income + (0.5 x Social Security benefits) + tax-exempt interest. The term “Modified Adjusted Gross Income” is used in various places in the tax code, but its definition varies with each use. Therefore, this study uses the term MAGIpi to denote this definition of MAGI. For most taxpayers, MAGIpi includes everything in adjusted gross income except the taxable portion of Social Security benefits.1

The taxable portion of Social Security benefits is the minimum of three equations: (1) 85 percent of Social Security benefits; (2) 50 percent of Provisional Income between the two PI income threshold levels plus 85 percent of Provisional Income beyond the second income threshold level; and (3) 50 percent of benefits plus 85 percent of Provisional Income beyond the second income threshold level.

The first or second formula provides the minimum amount for all households except some singles, qualifying widow(er)s, and married couples filing separately that lived apart for the entire year with less than $4,500 in annual Social Security benefits and some married couples filing jointly with less than $6,000 in annual Social Security benefits. The first and second PI income-threshold levels are $32,000 and $44,000 for married couples filing jointly and $25,000 and $34,000 for singles, qualifying widow(er)s, and married couples filing separately. These income threshold levels are not indexed for inflation.

Therefore, for married couples filing jointly, no Social Security benefits are taxable if Provisional Income is below $32,000. For most such couples, if PI is above $32,000, then taxable benefits is the lesser of: (1) 50 cents for each dollar of PI between $32,000 and $44,000 plus 85 cents for each dollar above $44,000; or (2) 85 percent of Social Security benefits.

The taxation of Social Security benefits results in a substantial hump in a household’s marginal tax rate when their income is within a wide income range. To illustrate this hump, consider the Jones family, a married couple filing jointly that will receive $60,000 in Social Security benefits in 2019. Each partner was at least 65 at year-end 2019. So, their standard deduction will be $27,000, which is the sum of the usual standard deduction of $24,400 for 2019 for married couples under age 65 plus an additional $1,300 each for being at least 65 at year-end 2019. Their “other income” consists of MAGIpi plus tax-exempt interest. This study assumed that they live in an income-tax-free state.2

Table 1 presents the Jones family’s 2019 marginal tax rates for various levels of other income (with all amounts rounded to the nearest dollar).


The first $17,784 of other income is tax free. At other income of $17,784, the Jones’ Provisional Income is $47,784, [$17,784 + 0.5($60,000)]. So, $9,216 of their Social Security benefits are included in AGI, [0.5($44,000 – $32,000) + 0.85($47,784 – $44,000)]. Thus, their AGI is $27,000 [$17,784 + $9,216]. Their standard deduction is $27,000, so their taxable income is $0. As a result, their federal-alone marginal tax rate is 0 percent for the first $17,784 of income.

Each dollar of other income between $17,784 and $28,270 causes another $0.85 of Social Security benefits to be taxed because their PI already exceeds the second threshold amount. Therefore, each dollar of income in this range causes their taxable income to rise by $1.85. They are in the 10 percent tax bracket. So, their federal-alone marginal tax rate is 18.5 percent, [10 percent tax bracket x 1.85].

At other income of $28,270, their tax bracket rises to 12 percent. So, each dollar of income between $28,270 and $60,459 causes an extra $0.85 of Social Security benefits to be taxed, and their marginal tax rate is 22.2 percent, [12 percent tax bracket x 1.85].

At other income of $60,459, their tax bracket rises to 22 percent. So, each dollar of income between $60,459 and $66,941 causes an extra $0.85 of Social Security benefits to be taxed, and their marginal tax rate is 40.7 percent, [22 percent tax bracket x 1.85].

At other income of $66,941, 85 percent of their Social Security benefits are taxable, which is the maximum.

The income range within which a household’s federal-alone marginal tax rate is 150 percent or 185 percent of the tax bracket is called the tax torpedo. For the Jones family, the tax torpedo refers to the range of other income between $17,784 and $66,941. At the end of the tax torpedo, there is a sharp drop in the marginal tax rate. For other income between $66,941 and $144,400, the Jones’ marginal tax rate falls sharply back to their tax bracket of 22 percent. The taxation of Social Security benefits causes a severe hump in marginal tax rates for other income within the $17,784 to $66,941 income range. At the end of this income range, there is a sharp fall in the marginal tax rate.3

As explained in Reichenstein (2019), the most tax-efficient withdrawal strategy from a household’s financial portfolio is usually quite different for middle-income and higher-income households. Although a more precise definition is provided later, in general, higher-income households include those that: (1) if they follow the conventional wisdom withdrawal strategy, and once their required minimum distributions begin, they will have incomes that are too high to avoid paying taxes on less than 85 percent of their Social Security benefits; but (2) their incomes may be high enough to force them to pay higher Medicare premiums. (These income-based increases in Medicare premiums are discussed in the next section.) The conventional wisdom withdrawal strategy calls for withdrawing funds in the financial portfolio in retirement from taxable accounts until exhausted, then from tax-deferred accounts such as 401(k)s until exhausted, and then from Roth accounts until exhausted.

In contrast, middle-income households are those: (1) whose income, if they follow the conventional wisdom withdrawal strategy, generally places them within the tax torpedo; but (2) their income will be low enough that they will not have to pay income-based spikes in Medicare premiums.

This study describes strategies higher-income families can adopt now that can: (1) substantially lower their lifetime income taxes; and (2) substantially lower their lifetime Medicare premiums.

If the Jones family follows the conventional wisdom withdrawal strategy, they would need other income exceeding $66,941 to be considered a higher-income household in this study. If this $66,941 of other income comes entirely from withdrawals from tax-deferred accounts (TDAs), then this income level would support $115,217 of spending (the sum of $66,941 in other income plus $60,000 in Social Security benefits, less federal income taxes).

If some of the $66,941 of other income comes from tax-exempt interest or qualified dividends and capital gains, then this level of income would support an even higher level of spending. Therefore, higher-income households, as defined in this study, include several households with well above-average spending levels. However, these higher-income households often have levels of financial wealth that are below levels most financial advisers would consider “high net worth” or “high-income” households. Thus, this study adopts the term “higher-income” households instead of high-income households.

According to the TCJA of 2017, tax rates are scheduled to return to the higher tax rates of 2017 and their often less-favorable tax brackets after COLA adjustments in 2026. Therefore, it is important to know how the tax torpedo will be affected if we return, as scheduled, to higher tax rates in 2026.

If we return to higher tax rates in 2026, then the highest marginal tax rate will rise from 40.7 percent to 46.25 percent. The tax bracket near the end of the tax torpedo will rise from 22 percent to 25 percent, and each dollar of additional income will still cause an extra $0.85 of Social Security benefits to be taxed. As a result, the marginal tax rate will rise from 40.7 percent to 46.25 percent, [25 percent tax bracket x 1.85]. Moreover, the range of income subject to this highest marginal tax rate is expected to widen. In short, the tax torpedo is scheduled to get worse in 2026.

Furthermore, as Geisler (2017) showed, the taxation of Social Security benefits can cause the federal-alone top marginal tax rate for taxpayers to be 49.95 percent based on the 2018 through 2025 tax structure, or 55.5 percent based on the scheduled tax structure beginning in 2026. So, Table 1 understates the adverse effects of the tax torpedo for some households.

Based on the 2018 through 2025 tax structure, suppose a household’s income is at the top of the range where long-term capital gains and qualified dividends are tax free. If this household withdraws another $100 from their TDA, it could increase the taxable portion of Social Security benefits by $85. This would cause an extra $185 of ordinary income to be taxed at 12 percent, the tax bracket at this income level. In addition, it would push $185 of long-term capital gains and qualified dividends above the range where long-term capital gains and qualified dividends are tax free. So, this $185 would be taxed at the 15 percent tax rate. The net effect is this household’s federal taxes would increase by $49.95 ($185 x 12 percent on ordinary income + $185 x 15 percent on long-term gains and qualified dividends). Its federal-alone marginal tax rate would be 49.95 percent.

Suppose we return, as scheduled, to the 2017 tax rates in 2026. Suppose a household’s income is at the top of the 15 percent tax bracket, which is also the income level below which long-term capital gains and qualified dividends are tax free. If this household withdraws another $100 from their TDA, it could increase the taxable portion of Social Security benefits by $85. This would cause an extra $185 of ordinary income to be taxed at 15 percent. In addition, it would push $185 of long-term capital gains and qualified dividends above the 15 percent tax bracket. Long-term gains and qualified dividends are stacked on top of ordinary income when calculating taxes. So, the last $185 of income—that is, the amount above the 15 percent tax bracket— would ​now be these long-term gains and qualified dividends. Therefore, this $185 would now be taxed at the preferential 15 percent tax rate. The net effect is this household’s federal taxes would increase by $55.50 ($185 x 15 percent on ordinary income + $185 x 15 percent on long-term gains or qualified dividends).

This study emphasizes the 46.25 percent and 40.70 percent federal-alone marginal tax rates caused by the taxation of Social Security benefits because they are more common, while acknowledging the potentially higher marginal tax rates explained by Geisler (2017). In short, even though the 46.25 and 40.70 percent marginal tax rates emphasized in this study are high, they understate the maximum federal-alone marginal tax rate some taxpayers will pay.

Income-Based Increases in Medicare Part B and D Premiums

The Affordable Care Act instituted higher Medicare premiums for retirees as their income level increases. The term denoting the portion of premiums that increases with income is called the Income-Related Monthly Adjustment Amount (IRMAA). In general, Medicare premiums for one year are based on Modified Adjusted Gross Income levels from two years earlier. This study used MAGImed to denote the definition of Modified Adjusted Gross Income as used to determine the level of Medicare premiums. MAGImed consists of adjusted gross income plus tax-exempt interest.

Table 2 shows how monthly 2019 Medicare premium levels increase when MAGImed levels in 2017 breach income threshold levels. The IRMAA premiums typically increase each year. The income threshold levels have been fixed from 2011 through 2019, but they are scheduled to increase with inflation beginning in 2020. The 2019 income threshold levels for MAGImed are $85,000, $107,000, $133,500, $160,000, and $500,000 for single taxpayers and $170,000, $214,000, $267,000, $320,000, and $750,000 for married couples filing jointly.


This study explained how 2017 MAGImed levels affect 2019 monthly Medicare premium for a single taxpayer, but the same logic applies to married couples. For MAGImed of $85,000 or lower, the standard premium for Part B applies, which in 2019 is $135.50 per month, and, when applicable, the plan premium for Part D (drugs) applies. The monthly Part D plan premium varies with each insurance plan. Part D coverage is optional. However, if MAGImed in 2017 exceeds $85,000 by a single dollar, then the monthly Part B premium in 2019 increases by $54.10 per month, while the Part D monthly premium, if applicable, increases by $12.40 per month.4

If a single taxpayer is covered by Part B, but not Part D, for all 12 months of 2019 then her annual premium rises by $649.20, [$54.10 x 12 months]. This is precisely like a $649.20 spike in income taxes for the $1 increase in income because she must pay this additional amount to the federal government. These IRMAA premium increases represent spikes in marginal tax rates. For example, this $1 rise in her income above $85,000 caused a spike in her marginal tax rate of 64,920 percent.

For a single individual enrolled in Parts B and D for all 12 months of 2019, the spike in Medicare premiums at income threshold levels of $107,000, $133,500, and $160,000 is about $1,210. For example, the jump in annual Parts B and D premiums at $107,000 is $1,209.60, [{($135.40 – $54.10) + ($31.90 – $12.40)} x 12 months]. The 2019 annual premium increases for married couples filing jointly are twice these amounts. Thus, a couple may have to pay more than $2,400 in additional joint annual Part B and D premiums in 2019 because their 2017 MAGImed was $1 above $214,000, $267,000, or $320,000. These IRMAA premium increases effectively represent spikes in their marginal tax rate that exceed 240,000 percent.5

The following case studies show the potential bene​fit to higher-income households of making Roth conversions in early retirement years while tax rates are scheduled to be temporarily lower.

Case Study No. 1: Higher-Income Married Couple

This study defines higher-income households as those that, if they follow the conventional wisdom (CW) withdrawal strategy, then once their required minimum distributions begin, they will have incomes that are too high to avoid paying taxes on less than 85 percent of their Social Security benefits.

However, these higher-income clients should be concerned with steps they can take before RMDs begin and before the temporarily lower tax rates expire, as scheduled, at the end of 2025 to lower: (1) the size of their lifetime income taxes; and (2) the size of their lifetime Medicare premiums. As this study’s two cases illustrate, these higher-income clients should consider making partial Roth conversions while tax rates are temporarily relatively low.6

The SECURE Act of December 2019 changed the beginning age for RMDs from 70½ to 72. Since the case studies presented here were written before this act was passed, they used 70½ as the beginning age for RMDs. Thus, the specific strategies in these two cases, which begin RMDs before they need to begin, are legal. Alternative strategies that delay RMDs until age 72 may be even better, and the software at incomesolver.com has been updated to reflect the new age to begin RMDs. However, the key lesson of this study remains the same: higher-income clients should consider making Roth conversions before RMDs begin and before higher tax rates are scheduled to return in 2026.

Jose and Maria Lopez are a married couple filing jointly. Jose was born Dec. 2, 1953. He has a Primary Insurance Amount (PIA) of $2,400 and a life expectancy of 80 years. Maria was born April 2, 1954. She has a PIA of $1,000 and a life expectancy of 95 years. They live in an income-tax-free state.

This case is run as if today is Jan. 1, 2020. They will spend $11,500 per month in real terms beginning in January 2020 with this spending level not including the cost of Medicare premiums. Maria will spend 80 percent of this amount after Jose’s death. They are both on Medicare Part B.

They have $500,000 in a joint taxable account (with cost basis of $500,000). Jose has $1 million in a TDA. Maria has $500,000 in a TDA. They maintain a balanced portfolio containing 56.3 percent fixed assets, including cash.

The federal tax code is the one established under the TCJA of 2017. Inflation will be 2.8 percent per year, and this inflation rate will affect their spending level, tax brackets, Medicare premiums, Social Security benefits, and the Medicare premium threshold levels.

Table 3 summarizes the results from three of their Social Security claiming and withdrawal strategies based on calculations by the Income Solver software.


Early strategy. In the conventional wisdom Social Security early strategy (CW SS Early), Jose files for his Social Security benefits in January 2020, and Maria files for her benefits the same month. After Jose’s death, Maria continues to receive Jose’s higher monthly benefits. They follow the conventional wisdom withdrawal strategy. Their portfolio lasts 27 years, but it fails to meet Maria’s spending needs in her last three years of life.

Primary strategy. In the conventional wisdom Social Security primary strategy (CW SS Primary), Maria files for her retirement benefits in January 2020, and Jose files a restricted application for spousal benefits that same month of $500 per month (half of Maria’s PIA). At age 70, Jose files for his retirement benefits and Maria adds spousal benefits. After Jose’s death, Maria continues to receive Jose’s higher monthly benefits. They follow the conventional wisdom withdrawal strategy. Their portfolio lasts 28 years, but it fails to meet Maria’s spending needs in her last two years of life.

Best strategy. In the “best” strategy, they follow the primary Social Security claiming strategy. Each year from 2020 through 2022, they withdraw funds from their financial portfolio following the conventional wisdom withdrawal strategy to meet their spending needs, and then convert funds from their TDA to a Roth IRA to fill the 22 percent tax bracket.

In 2023 through 2025, they withdraw enough funds from their TDA to meet their spending needs, which keeps them in the 22 percent tax bracket. From 2026 (when higher tax rates return) through 2047, they withdraw enough funds from their TDAs to fill the 15 percent tax bracket and then withdraw additional funds to meet their spending needs from their Roth IRAs; their taxable account has already been exhausted. Maria’s Roth IRA is exhausted in 2048, and her tax bracket is 25 percent in 2048 and 2049. Their portfolio lasts this couple’s entire joint lifetime of 30 years and an ending balance of $114,242 remains for their heirs.7

Comparing the strategies. A smart Social Security claiming strategy allows this couple to extend the longevity of their financial portfolio by one year. Combining a smart Social Security claiming strategy with a tax-efficient withdrawal strategy extends the longevity of their portfolio by three years—through both of their lifetimes—and funds remain in their financial portfolio for their heirs. This combination of a smart Social Security claiming strategy and a tax-efficient withdrawal strategy added $578,511 in total value to the family.8

The Social Security claiming strategy in the CW SS primary strategy provides $260,936 more in nominal lifetime benefits than the early claiming strategy. From 2020 through 2023, the early claiming strategy provides more Social Security benefits, but the primary claiming strategy provides about $10,000 more in annual benefits in 2024 with this amount rising with inflation in subsequent years.

Partially offsetting the primary claiming strategy’s relative advantage in terms of nominal lifetime benefits is the fact that the larger Social Security benefits in 2020–2023 with the early claiming strategy occur in years when they pay no income taxes. In these early years, they withdraw funds from their taxable account, and these funds are largely tax-free withdrawals of principal. Altogether, the primary claiming strategy increased the projected total value of the Lopez portfolio by $120,252 compared to the early claiming strategy.

In the CW SS primary strategy, this couple is in the 0 percent tax bracket from 2020 through 2023, because funds to meet their spending needs are being withdrawn entirely from their taxable account. Both partners begin RMDs in 2024, and they are in the top end of the 12 percent tax bracket in 2024 and low end of the 22 percent tax bracket in 2025. From 2026 (when the higher tax rates return) through 2033 (the year Jose dies) they are in the 25 percent tax bracket.

From 2034 through 2047, Maria is in the 28 percent tax bracket. Her portfolio is exhausted in 2028. She is in the 10 percent tax bracket in 2048 and 0 percent tax bracket in 2049. From 2024 (the year Jose’s age-70 benefits level begins) through 2047, they pay taxes on 85 percent of Social Security benefits. Therefore, they are a higher-income household as defined in this study.

Due to the taxation of Social Security benefits, in 2025 they pay a marginal tax rate of 40.7 percent on some of their TDA withdrawals. In 2026 through 2047 they (or Maria after Jose’s death) pay a marginal tax rate of 46.25 percent on some of their TDA withdrawals.

In the best strategy, from 2020 through 2025, they are in the 22 percent tax bracket. From 2027 through 2047, they are in the top end of the 15 percent tax bracket. Maria’s Roth IRA is exhausted in 2048, which raises her tax bracket to 28 percent in her last two years.

Recall that in the CW SS primary strategy they are in the 0 percent tax bracket from 2020 through 2023. In the best strategy they do not waste the opportunity to convert pre-tax funds in TDAs to after-tax funds in Roth IRAs in these years at 0 percent through 22 percent tax rates. This is far better than having some of these TDA withdrawals eventually taxed at marginal tax rates of 40.7 percent or 46.25 percent in the CW SS primary strategy.

Figure 1 illustrates the pattern of taxes with the best and CW SS primary strategies. In the best strategy, they pay higher taxes from 2019 through 2025 with this withdrawal strategy. In 2026 and 2027, after making TDA withdrawals to the top of the 15 percent tax bracket, they withdraw funds from their taxable account to meet the rest of their spending needs. Beginning in 2028, after making TDA withdrawals to the top of the 15 percent tax bracket, they (or Maria after Jose’s death) make tax-free withdrawals from their Roth IRAs to meet the rest of their spending needs. This allows them to pay lower taxes with the best strategy in 2026 through 2047.


For example, in 2034 Maria’s taxes are projected to be $8,223 in the best strategy, while her taxes are projected to be $37,400 in the CW SS primary strategy. In the CW SS primary strategy, all withdrawals from her financial portfolio in 2034 come from her TDA and some of these withdrawals are taxed at a marginal tax rate of 46.25 percent due to the taxation of Social Security benefits.

In contrast, with the best strategy, they remain in the 15 percent tax bracket until the Roth IRA is exhausted in 2048. This allows them (or Maria after Jose’s death) to avoid withdrawing additional TDA funds that would be taxed at 46.25 percent. Instead, in these years, they (or Maria after Jose’s death) pay a marginal tax rate of 27.25 percent, [15 percent tax bracket x 1.85], on some of their TDA withdrawals.

Furthermore, from 2034—the first year Maria files as a single taxpayer—through 2047, she pays taxes on less than 85 percent of her Social Security benefits with the best strategy. Altogether, this couple is projected to pay $435,519 less in lifetime income taxes with the best strategy compared to the CW SS primary strategy.

Separately, compared to the CW SS primary strategy, in the best strategy this couple pays higher Medicare premiums in 2022 through 2024 (due to their higher incomes in 2020 through 2022), but Maria will pay lower Medicare premiums beginning in 2036 (due to her lower income beginning in 2034, the first year she files as a single individual) and through the rest of her life.

For example, in 2034 Maria’s MAGI-med with the CW SS primary strategy is projected to be $191,108, including $47,558 in taxable Social Security benefits, while with the best strategy, her MAGImed is projecte​d to be $79,987, including $36,775 of taxable Social Security benefits. The lesson is that the early-year Roth conversions, while both partners are alive, can dramatically reduce the survivor’s income after the death of the first spouse, which can substantially reduce the survivor’s Medicare premiums. Remember, the first four Medicare income threshold levels for singles are at half the levels as for married couples filing jointly. Altogether, they are projected to pay $35,356 less in lifetime Medicare premiums with the best strategy.

In short, based on the current tax code as legislated in the TCJA, this couple can extend the longevity of their portfolio through both of their lifetimes by making Roth conversions to the top of the 22 percent tax bracket from 2020 through 2022. These early-year Roth conversions allow them to greatly reduce not only their lifetime taxes, but also their lifetime Medicare premiums.

Case Study No. 2: Higher-Income Single Household

Sarah Jones is single. She was born Dec. 2, 1955. She has a Primary Insurance Amount (PIA) of $2,600 and a life expectancy of 93 years. She lives in an income-tax-free state. This case was run as if today is January 1, 2020.

She will spend $9,500 per month in real terms beginning in January 2020, with this spending level not including the cost of Medicare premiums. At age 65, she will file for Medicare Part B and Part D with a plan premium of $40 per month. She has $500,000 in a taxable account (with cost basis of $500,000) and $1.5 million in a TDA. She maintains a portfolio that contains 40 percent stock and 60 percent bonds, including cash.

The federal tax code is the one established under the TCJA. Inflation will be 2 percent per year, and this inflation rate will affect her spending level, tax brackets, Medicare premiums, Social Security benefits, and the Medicare premium threshold levels. Her beneficiaries are assumed to pay, on average, 10 percent of TDA withdrawals in taxes.

Table 4 summarizes the results from three of her Social Security claiming and withdrawal strategies based on the calculations by the Income Solver software.


Early strategy. In the conventional wisdom early strategy (CW SS Early), Sarah files for Social Security retirement benefits in January 2020 and she follows the conventional wisdom withdrawal strategy. Her portfolio lasts 26 years, but it fails to meet her spending needs in her last three years of life.

Primary strategy. In the conventional wisdom primary strategy (CW SS Primary), Sarah files for Social Security retirement benefits at age 70 and she follows the conventional wisdom withdrawal strategy. Her portfolio lasts 26 years, but it fails to meet her spending needs in her last three years of life.

Select strategy. In the “select” strategy, she files for Social Security retirement benefits at age 70. From 2020 through 2024, she withdraws funds following the conventional wisdom withdrawal strategy to meet her spending needs and then converts funds from her TDA to a Roth IRA to fill the 24 percent tax bracket. In these years, she is projected to convert $713,695 of funds to her Roth IRA.

In 2025, she withdraws funds from her TDA to the top of the 22 percent tax bracket and then withdraws additional tax-free funds from her Roth IRA to meet the rest of her spending needs. From 2026 (when tax rates rise) through 2046, she withdraws from her TDA the larger of: (1) RMDs; or (2) the amount of funds to fill the top of the 15 percent tax bracket, which is always the amount of funds to fill the 15 percent tax bracket. Her Roth IRA is exhausted in 2047 and she is in the 28 percent tax bracket in 2047 and 2048. Her portfolio lasts her lifetime of 29 years and there is an ending balance of $49,441 that remains for her heirs.

In summary, a smart Social Security claiming strategy adds about $77,000 in total value. The select strategy’s tax-efficient withdrawal strategy adds three years of longevity and adds about $441,000 in additional total value to Sarah’s portfolio.

Figure 2 presents the taxes paid by year in the select and CW SS primary strategies. With the select strategy, Sarah pays more in taxes from 2020 through 2024. However, the ability to withdraw tax-free funds from her Roth IRA in 2026 through 2046 keeps her income at the top of the 15 percent tax bracket. In addition, with the select strategy, Sarah pays taxes on less than 85 percent of her Social Security benefits from 2026 through 2046.


In contrast, in the CW SS primary strategy, Sarah is in the 0 percent tax bracket from 2020 through 2023 when funds are being withdrawn exclusively from her taxable account. She is in the 10 percent tax bracket in 2024, the year her taxable account is exhausted. Thus, in these early retirement years, she wastes the opportunity to convert pre-tax funds in her TDA to after-tax funds in a Roth IRA at marginal tax rates of 0 percent through 24 percent. Instead, in this strategy, some of these TDA funds are eventually taxed at 40.7 percent or 46.25 percent marginal tax rates.

In the CW SS primary strategy, her tax bracket is 24 percent in 2025. From 2026 through 2045, Sarah’s income places her in the 28 percent tax bracket. Her financial portfolio is exhausted in 2046 and she is in the 25 percent tax bracket that year. She is in the 0 percent tax bracket in her last two years of life. In addition, she pays taxes on 85 percent of her Social Security benefits every year until her portfolio is exhausted in 2046. Altogether, Sarah is projected to pay $385,925 more in lifetime federal income taxes with the CW SS primary strategy.9

In the select strategy, Sarah pays higher Medicare premiums in 2022 through 2026 due to her higher income in 2020 through 2024 than in the CW primary strategy. However, she pays lower Medicare premiums in 2027 through 2047, the year before she dies. Altogether, Sarah is projected to pay $65,819 less in Medicare premiums with the select strategy.

In short, by using Roth conversions in the early retirement years, this higher-income single taxpayer was able to greatly reduce both her lifetime taxes and her lifetime Medicare premiums.

Implications for Financial Planners

This study presented two cases that demonstrated that financial planners can add considerable value to higher-income households by helping them coordinate a smart Social Security claiming decision with a tax-efficient withdrawal strategy.

These two cases considered higher-income households that are younger than the age when RMDs begin. In general, these households are those with incomes that are too high to avoid paying taxes on less than 85 percent of Social Security benefits if they follow the conventional wisdom withdrawal strategy, but their incomes are high enough that they should consider how the withdrawal strategy from their financial portfolio will affect the size of their Medicare premiums.

These cases demonstrated that these higher-income households should consider Roth conversions before RMDs begin and before tax rates are scheduled to increase in 2026. These early-retirement-year Roth conversions can greatly decrease both their lifetime income taxes and their lifetime Medicare premiums compared to the levels they would pay if they followed the conventional wisdom withdrawal strategy.

For higher-income households that follow the conventional wisdom withdrawal strategy, in their first few years of retirement, when funds are being withdrawn from their taxable account and before required minimum distributions have begun, they will be in low tax brackets, possibly 0 percent. However, during most of their retirement years, when funds are being withdrawn from their TDAs, they will be in the 25 percent or higher tax bracket. These higher-income households that follow the conventional wisdom withdrawal strategy waste the opportunity in these first few retirement years to convert pre-tax funds in their TDAs to after-tax funds in Roth IRAs at low tax rates. Instead, most of these TDA balances are withdrawn in 2026 or later when they will pay marginal tax rates, due to the taxation of Social Security benefits, of up to 46.25 percent on some of these TDA withdrawals.

In contrast, in the recommended withdrawal strategies that use Roth conversions, these retirees do not waste the opportunity to convert funds to a Roth IRA to fill the 0 percent, 10 percent, 12 percent, 22 percent, and, in one of the two cases, the 24 percent tax brackets in the early retirement years when funds to meet their spending needs are being withdrawn primarily, if not entirely, from taxable accounts and before required minimum distributions have begun. The marginal tax rate on these converted funds is the same as the tax bracket, that is, 0 percent to 24 percent. These marginal tax rates are much lower than the 46.25 percent marginal tax rates eventually paid on some of these TDA withdrawals beginning in 2026 by retirees that follow the conventional wisdom withdrawal strategy.

As these cases demonstrate, it is far better to make Roth conversions in the first few retirement years and pay marginal tax rates of 0 percent to 24 percent on these TDA balances that are converted to a Roth IRA than to not make these conversions and have these TDA withdrawals eventually taxed at up to 46.25 percent.

These higher-income households that use Roth conversions in their early retirement years can then withdraw tax-free funds from their Roth IRAs in later retirement years. These later retirement years begin when higher tax rates return in 2026, if not sooner. This ability to withdraw tax-free funds from their Roth IRAs: (1) keeps them in relatively low tax brackets; (2) allows them to pay taxes on less than 85 percent of their Social Security benefits; and (3) significantly reduces their Medicare premiums.

The reduction in annual Medicare premiums can be especially strong for the surviving spouse three years after the death of the first spouse, because the survivor will have to face Medicare income threshold levels that are generally half the size of the income threshold levels facing married couples.

In short, these cases demonstrate that higher-income retirees who use Roth conversions before RMDs begin and before tax rates are scheduled to increase in 2026 are likely to pay: (1) significantly lower lifetime income; and (2) significantly lower lifetime Medicare premiums than retirees who adopt the conventional wisdom withdrawal strategy.

Endnotes

  1. ​For 2018, MAGIpi includes amounts in wages, salaries, tips, etc.; taxable interest; ordinary dividends; taxable amount of IRA, pension, and annuities; and Line 22 from Schedule 1. It then excludes amounts from Lines 23 through 32 of Schedule 1, but it does not exclude Line 33, student loan interest deduction. Since few retirees have student loan interest deductions, MAGIpi is generally everything in adjusted gross income except the taxable portion of Social Security benefits.
  2. If they live in a state that imposes an income tax, then the relative advantage of the strategies recommended in this study will usually be larger.
  3. The tax torpedo refers to the “other income” range between the level where taxable income is $0 and the level where 85 percent of Social Security benefits are taxed. For all single households and married households receiving $45,000 or less in annual Social Security benefits, the marginal tax rate within the tax torpedo is either 150 percent or 185 percent of the tax bracket. For the married couple in this example with $60,000 in annual Social Security benefits, the level of other income that corresponds to taxable income of $0 already produces a level of Provisional Income that places them beyond the second income threshold level of $44,000. So, they have no income range where their marginal tax rate is 150 percent of their tax bracket.
  4. The increase in Medicare premiums occurs if MAGImed exceeds an income threshold level by $0.01. However, because marginal tax rates are defined as the additional taxes paid on the next dollar of income, the impact if a household’s MAGImed exceeds an income threshold level by $1 is discussed in this paper.
  5. In general, Medicare premiums for one year are determined by the tax filing status and the MAGImed level from two years earlier. There are exceptions to this general rule. If a life-changing event (LCE) would lower a client’s MAGImed sufficiently to lower his or her IRMAA premiums for a specific year, then he or she can file Form SSA-44 to report this LCE. LCEs include divorce/annulment, death of a spouse, work stoppage or work reduction, loss of income-producing property due to a disaster, loss of pension income, and situations where income for a year was due to a settlement with an employer for the employer’s bankruptcy or reorganization. For more information on Medicare premiums and IRMAAs, see the Social Security Administration’s “Medicare Premiums: Rules for Higher-Income Beneficiaries,” at ssa.gov/pubs/EN-05-10536.pdf​ and “Medicare Income-Related Monthly Adjustment Amount: Life-Changing Event,” at ssa.gov/forms/ssa-44-ext.pdf​.
  6. Separately, once these clients reach age 70½, they can meet any charitable intentions they may have by directing their contributions through qualified charitable distributions (QCD) from their IRA. The case studies here concentrate on the first step. The bottom line is that donating money through a QCD from an IRA will reduce a household’s MAGImed compared to the traditional contribution method of distributing funds from a tax-deferred account, depositing these funds in a checking account and then writing a check for this amount to the qualified charity. For more information, see Cook, Harelik, and Reichenstein (2016) and Gardner and Daff (2017).
  7. This ending balance is an after-tax amount. Their heirs are projected to inherit $142,803 from a TDA, but their beneficiaries are assumed to lose 20 percent of these pre-tax balances to taxes. Therefore, the after-tax value is $142,803 x (1 – 0.2) = $114,242.
  8. This total value is an after-tax amount. It is the sum of spending, which requires after-tax funds, plus the ending balance, if any, which is also an after-tax amount.
  9. It is worth noting that if Sarah dies at age 91, the select strategy would have an ending balance of $308,068. Her Roth IRA has a trivial balance when Sarah turns 91 in December 2046. Thus, virtually all withdrawals in 2047 and 2048 come from her TDA, which greatly reduces the ending balance that goes to her heirs.

References

Blanchett, David, and Paul Kaplan. 2013. “Alpha, Beta, and Now…Gamma.” The Journal of Retirement 1 (2): 29–45.

Cook, Kirsten A., Harry Harelik, and William Reichenstein. 2016. “The Tax Advantages of Qualified Charitable Distributions from IRAs.” AAII Journal 38 (10): 25–28.

Gardner, Randy and Leslie Daff. 2017. “How to Use Qualified Charitable Distributions as a Tax-Saving Tool.” Journal of Financial Planning 30 (8): 30–32.

Geisler, Greg. 2017. “Taxable Social Security Benefits and High Marginal Tax Rates.” Journal of Financial Service Professionals 71 (5): 56–67.

Kinniry, Francis M. Jr., Colleen M. Jaconetti, Michael A. DiJoseph, Yan Zilbering, and Donald G. Bennyhoff. 2016. “Putting a Value on Your Value: Quantifying Vanguard Advisor’s Alpha.” Available at vanguard.com/pdf/ISGQVAA.pdf.

Reichenstein, William. 2019. Income Strategies. Leawood, Kansas. Retiree Income Inc.

Reichenstein, William, and William Meyer. 2018. “Understanding the Tax Torpedo and Its Implications for Various Retirees.”​ Journal of Financial Planning 31 (7): 38–45.

Citation

Reichenstein, William, and William Meyer. 2020. “Using Roth Conversions to Add Value to Higher-Income Retirees’ Financial Portfolios.” Journal of Financial Planning 33 (2): 46–55.

Member Access

Includes:
Current Issue
Digital Edition
CE Exams
Supplements
Podcasts

Subscribe

Change Address