The Neglected Secret to Tax-Efficiently Utilize IRA Basis

Your tax return is nearly complete and your tax accountant or the tax filing software you are using informs you that you can still make an IRA contribution for the previous tax year. You remember something you read in the past from a reputable source encouraging IRA contributions as a wise tax move. You have the cash available and elect to make the annual IRA contribution. Seems like a good idea that might save taxes in the future. Furthermore, you repeat the same process in years to follow.

It was all well-intentioned. And this is how a potential tax problem begins. Most married taxpayers with six-figure income who are eligible to participate in an employer-sponsored retirement plan (such as a 401k or 403b) are not permitted to make deductible IRA contributions. The income limits for single filers are even lower (income eligibility figures can be found here). When you make an IRA contribution but are not eligible for a deduction, the funds are treated as an after-tax contribution (also called a non-deductible contribution). The contribution is then deemed to be “IRA basis” and has to be diligently tracked every year thereafter via the IRS Form 8606 to avoid being taxed twice. We’ll explain this in a moment.

As a general rule, we discourage clients from making these non-deductible IRA contributions unless the contribution is specifically intended for a backdoor Roth strategy. The biggest problem with non-deductible IRA contributions (those not intended for a backdoor Roth), as we explain in detail here, is that the contributions usually result in higher taxes – often because the contributions are not tracked properly over the course of decades and get taxed twice or simply because the subsequent gains get taxed at higher rates than they would otherwise be taxed.

But back to basis. If you made non-deductible IRA contributions in the past, you should be tracking the basis each year with the Form 8606 that accompanies your tax return. Each time you make a new non-deductible IRA contribution, you add to the basis. Not diligently tracking this basis results in a costly problem – paying excessive future taxes because of double taxation. Happens all the time. The error of not properly tracking basis can and should be fixed by correcting the Form 8606 dating back to the initial non-deductible IRA contribution. The more non-deductible contributions you have made without proper tracking, the bigger the adverse tax problem you will face – if not corrected.

Assume now that you are properly tracking this basis. This means that some portion of your retirement account balance will not be subject to taxation when eventually distributed. Consider the following example to help understand how this works:

Example 1a: Jill made non-deductible IRA contributions for each of the past six years (2015-2020) because her income and participation in a workplace 401k prohibited her from making deductible IRA contributions. In total, her non-deductible IRA contributions equal $34,000 ($5,500 limit for 2015-2018, $6,000 for 2019-2020). This $34,000 is Jill’s IRA basis.

Jill had a pre-existing IRA balance before 2015 (funded entirely with deductible contributions). The 2015-2020 non-deductible contributions combined with the pre-existing balance and subsequent growth brings the total current value of her IRA to $207,000. This means that 16.43% of her IRA ($34,000 / $207,000) has already been subjected to taxes and will avoid taxes in the future. The remaining 83.57% is subject to future taxes. Each dollar that comes out gets taxed based on these percentages. If Jill were to distribute $100 right now, $83.57 would flow through to her tax return as income subject to taxation. Had she not kept track of the basis, she would pay taxes on the entire $100 distribution.

What if Jill wants to do a partial Roth conversion for $34,000? The same math applies – 83.47% of the amount she converts will be subject to taxation ($28,415 of the $34,000). The basis is considered cream in the coffee. Once the cream gets mixed with the coffee, any distributions are pro rata distributions for tax purposes.

There is, however, a widely neglected or unknown workaround that allows for separating the cream and the coffee. This workaround – if properly executed – allows Jill to convert the full $34,000 and use only basis in the conversion. As a result, none of the $34,000 conversion is taxable and she has converted all $34,000 of basis to her Roth without triggering any taxes. This is a big opportunity for anyone with IRA basis.

The strategy only works if Jill participates in a 401k that accepts rollover contributions (the good news is that most 401k plans allow for inbound rollovers). The reason it works is that 401k plans are prohibited from accepting basis with rollover contributions. The IRS specifically disallows basis on these IRA to 401k rollovers such that a rollover must consist entirely of pre-tax dollars. As a result, someone in Jill’s situation is actually able to isolate the basis. Continuing the example:

Example 1b: Jill desires to convert the $34,000 of basis in her IRA to her Roth IRA but learns that doing so will result in 83.47% of the distribution being taxable. Her tax advisor informs her that because she participates in a workplace 401k plan that accepts rollover contributions, she can rollover $173,000 from her IRA to her 401k ($207,000 less the $34,000 basis). Because the 401k is not permitted to accept basis, the entire $173,000 rollover comes from pre-tax IRA dollars which then isolates the remaining $34,000 in her IRA as 100% basis. A few days after rolling the $173,000 to her 401k, she converts the $34,000 remaining in her IRA to a Roth IRA. Because the IRA consisted entirely of after-tax basis, the conversion results in zero taxable income and zero tax cost.

While this works beautifully for Jill, what happens for someone in the same situation without access to a 401k that accepts rollovers? Consider the following example:

Example 2a: Michael retired several years ago from his job as an actuary and has a large IRA worth $2,000,000. He diligently tracked his after-tax IRA contributions and has basis of $90,000 within the IRA that gets reported each year on his Form 8606. He would like to separate the basis dollars for a tax-free Roth conversion but does not participate in a workplace 401k plan that would allow him to take advantage of this Roth conversion strategy.

Once or twice a year, Michael gets paid a small fee to speak at actuarial conferences – usually a few hundred dollars. His financial planner suggests that Michael establish an EIN (business tax ID) at no cost and open an individual 401k (also no cost) to save some of the income from his speaking engagements. While the tax deferral benefit will be minimal as Michael earned just $650 from one engagement this year, the individual 401k will provide an opportunity for Michael to isolate the basis from his IRA. Michael opens the individual 401k as advised and then converts $1,910,000 from his existing IRA to the new 401k. This leaves $90,000 remaining in his IRA which is 100% after-tax basis. Immediately thereafter, Michael converts the remaining $90,000 IRA to a Roth IRA at zero cost.

Granted, most retirees may not have speaking engagements once they retire. But the opportunity to use the same strategy as Michael is far-reaching. The reality is that all it takes to establish an EIN is to get paid $15 to watch the neighbor’s dog for a day. Or sew a few facemasks and sell them on Etsy. Or organize a buddies’ golf trip and collect a few dollars from each participant as an organization fee. There are countless possibilities to earn a few dollars and legitimately use the “job” as a way to establish an individual 401k for the basis conversion workaround strategy. Moreover, you don’t have to wait until retirement to start a side job that allows for legitimately opening an individual 401k.

Notably, there is one final important caveat to this strategy that, if ignored, could destroy the entire design. For the strategy to work effectively, neither Jill nor Michael from the prior examples should rollover any funds from their 401k plans back to a Traditional IRA until at least the next calendar year. Doing so would cause any funds rolled back to the IRA to be included in the calculation of the Roth conversion taxable amount, even though it was rolled over back to an IRA after the conversion. The following extension of Michael’s example explains:

Example 2b: Continuing from Michael’s example above, assume that he converts the $90,000 IRA to a Roth IRA in June and decides immediately thereafter that he has no more use for the individual 401k and does not want to be bothered with it. Two months later – In August – he rolls over the $1,910,000 from the Individual 401k back to a new IRA and closes down the Individual 401k. By rolling these funds back to an IRA in the same calendar year, the taxable portion of his $90,000 Roth conversion now has to proportionately incorporate the $1,910,000 on the Form 8606. That is – the after-tax portion of the $90,000 Roth conversion that he thought was successfully achieved without any tax impact is no longer 100% ($90,000/$90,000). Instead, the after-tax portion is now only 4.5% ($90,000/$2,000,000). As a result of this miscue, Michael will have to report $85,950 of taxable income on the conversion.

Closing Comments

This may all sound really wonky and of limited impact. It may be wonky but it is not of limited impact. We find that roughly half of the prospective clients we speak with have made non-deductible IRA contributions in the past. Generally speaking, they made IRA contributions when they were earning lower incomes and eligible to deduct the contributions and then continued to make contributions after their income started to exceed the deductibility thresholds. Some are properly accounting for this basis each year while others have lost track of the basis.

In all of these scenarios, the basis isolation approach should be immediately considered because the sooner the basis is isolated and converted to a Roth IRA, the more lifetime taxes will be avoided. In Michael’s example, converting the $90,000 to Roth right now means that any future growth on the $90,000 also avoids future taxation. Were Michael simply to ignore the opportunity and the IRA subsequently doubled in value over the next decade, he would owe tax on $90,000 of growth that could have been avoided1. And this is a very important takeaway: the sooner that IRA basis is isolated and converted to Roth as outlined above, the greater the future tax savings will be2.

 

After-Tax IRA Contributions Are Often a Financial Mistake

High income families faced with high annual taxes are often looking for ways to reduce their taxes or save additional dollars in tax-efficient ways.  Unfortunately, the pursuit of lower taxes often results in irrational behaviors or unfavorable outcomes (such as paying more taxes than you would have otherwise paid by doing nothing).  Among the classic examples of generally ill-advised tax-saving strategies are non-deductible IRA contributions.

What is a Non-Deductible IRA Contribution?

Different rules apply for making deductible IRA contributions depending on marriage status, income, and eligibility for a workplace retirement plan.  Individuals not covered by a workplace retirement plan (such as a 401k) or married couples where neither spouse is covered by a workplace plan are eligible to make IRA contributions, regardless of income.  Alternatively, a husband and a wife who are both covered at work by retirement plans cannot make deductible IRA contributions if their combined income exceeds $124,000 (2020).  The full set of deductible contribution limits can be found here.

Many successful working professionals eventually become disqualified to make deductible IRA contributions as their income grows.  Despite pervasive public belief to the contrary, this is not the end of the road for IRA contributions.  The IRA deduction rules merely limit the ability of individuals to make deductible IRA contributions – they do not prevent individuals from making non-deductible IRA contributions.  In fact, any working individual or spouse of a working individual, regardless of income or availability of a workplace retirement plan, is eligible to make non-deductible IRA contributions.

The basic idea of a traditional IRA is that you make contributions each year which can then be deducted from income and resultantly reduce your tax liability.  The tax is avoided upfront but paid in the future – ideally after many years of tax-deferred growth.

If the benefit of a traditional IRA contribution is the immediate tax deduction, then why contribute after-tax dollars to an IRA without any deduction?  There are generally two viable reasons to make non-deductible IRA contributions.  First, the individual may be a prime candidate for the backdoor Roth conversion – the initial step of which is the non-deductible IRA contribution.  The second reason is that any growth or income that accrues inside the IRA after the contribution is tax-deferred and only taxed whenever distributions begin in the future.

What Are the Drawbacks of Non-Deductible IRA Contributions?

Given the potential benefit of deferred taxation, non-deductible IRA contribution may seem to be an underutilized opportunity.  However, there are a number of potential drawbacks that tend to make these contributions only useful for short-term traders.

1) Favorable capital gain tax rates instead become less favorable ordinary income tax rates.  

Non-deductible IRAs do not eliminate taxes on gains – they merely defer taxes.  In many cases, simply buying an index fund in a regular brokerage account can be far more tax efficient than buying the same fund in a non-deductible IRA.

Consider the scenario where you contributed $6,000 to a non-deductible IRA, invested in a stock fund, and the investment grew over the past 25 years at 8% per year.  You’re now ready to liquidate the IRA – now valued at $41,091.  Because of other income sources and Social Security, you find yourself in the 24% federal tax rate which means paying taxes of $8,421 on the distribution.  Had you made the same investment in a regular brokerage account, you would be paying long-term capital gain taxes at 15% which translates to $5,263 of taxes – a savings of more than $3,000.

If you’re planning to day-trade the investments or do a lot of trading and would not likely achieve the minimum 1-year holding period to qualify for long-term capital gain rates, then the deductible IRA would be a better choice.  Otherwise, the premise that the non-deductible IRA will save taxes over the long-run may be ill-conceived.

2) You face forced taxable distributions in the future.  

Even if you had not needed to liquidate the IRA account in the example above, the IRS would have forced you to start taking mandatory distributions beginning at age 72 and every year, thereafter.  Each of these distributions would again be taxable as the less favorable ordinary income rates.  In contrast, the regular brokerage account imposes no forced distributions so investments can be left to grow tax deferred until the funds are needed.

3) You lose the potentially valuable step-up in basis.

Continue with the same example and assume that you never needed to use the proceeds from the hypothetical investment account.  At death, the IRA would still be taxable as ordinary income to your beneficiaries at their then-current tax rate.  In contrast, the regular brokerage account gets a favorable “step-up in basis” at death meaning that your beneficiaries could sell the appreciated investments and owe zero taxes.  The non-deductible IRA contribution converted $35k of gains into ordinary income and $8.4k of taxes that could have entirely been avoided.

4) There’s a good chance you pay taxes on the same dollars, twice.  

In the world of fallible human beings, the best action is often not the economically optimal option.  That is, practical considerations matter and often matter a lot.  Someone could evaluate the tax benefits of making non-deductible IRA contributions and determine that they are economically optimal based on the individual circumstances.  Yet this still does not mean the IRA contributions are the best course of action.

When you make a non-deductible IRA contribution, the IRS expects that you file a Form 8606 not only in the year of the contribution but every year, thereafter.  This form tracks your IRA basis so that when it comes to distribute from the IRA, you’re not paying taxes on the same dollars twice.  In the real world, many people lose track of filing this form.  They switch to a new accountant or to new tax software and Form 8606 stopped getting filed.  Forgotten IRA basis is even more likely when the IRA owner dies.  Although beneficiaries are able to avoid tax on any inherited IRA basis, this information almost never gets communicated from the executor to the beneficiaries.  The next time a new client brings in an inherited IRA and knows the basis or received the Form 8606 from the executor will be the first time.

All of this simply means that a large amount of non-deductible IRA contributions are being taxed twice – once at the time of the contribution (since the contribution is made with after-tax dollars) and then at the time of the distribution (since without a record of basis, all distributions are assumed to be taxable).  Speaking from experience, we would bet that more IRA basis is ultimately lost and taxed twice than the amount properly recorded and taxed just once.  This likelihood of double taxation is another real-world drawback of the non-deductible IRA contribution as it contradicts with the initial objective of reducing taxes.

Closing Thoughts

Experience suggests that many high income earners start by looking for perceived tax-efficient saving vehicles like 401(k) plans, non-deductible IRAs, or tax deferred annuities and then fill up those buckets without really examining the ultimate tax efficiency of the entire retirement savings strategy.  We suggest a better approach is starting not with the solutions but with the question “how much do I need or want to save for retirement this year?”  Only after first addressing that question does it make sense to then evaluate the implementation.  In many cases, investors will find that the perception of tax savings from strategies like the non-deductible IRA contribution are really just perceptions that get in the way of a useful big picture retirement strategy.

Two Presidential Candidates Walk into a Bar

“It is a habit of mankind to entrust to careless hope what they long for, and to use sovereign reason to thrust aside what they do not fancy.” – Thucydides

Stop me if you have heard this before: The upcoming election is the most important election in our country’s history. A polarizing divide between candidates and policies means that the decision we make this November will reverberate for generations to come. Simply put, the future of our republic hinges on the outcome.

Such has been the rhetoric of candidates, media pundits, political commercials, and headlines starting in 1792 and repeating every four years thereafter. Whatever represented the 18th and 19th century equivalents of Twitter almost assuredly had “experts” loudly opining on the momentous importance of each election and the potentially grave historical consequences of “poor decisions”. 

This is not to suggest that the 2020 election is more or less historically important than past elections. Maybe this time is different and 2020 is the most important election in history. 

The point here is not to argue the historical significance of this election but to suggest that nearly all of what we so repeatedly consume each election cycle needs to be taken with a big grain of salt. This is no more true than in the purported connections between the election and Wall Street. While there are undeniably consequences to the economy and the stock market stemming from the election results, much of what we are led to believe about this connection is misleading, if not false. To put this in perspective, we outline several of the fictional narratives that – in some fashion – become mainstream thinking every four years. 

Fiction 1: Given the uncertainty surrounding the election, it is a good idea to hold cash and wait until after the election to invest.

Since investment markets move higher over time, this line of thinking inherently presupposes that markets will perform poorly leading up to a presidential election and that the clarity of a winner will then lead markets higher. Fact: the stock market has performed dramatically better in the months leading up to Presidential elections. Not just a little better – dramatically better. The chart below shows the combined return of the S&P 500 in the four months leading up to (presidential) election days starting with the 1928 election compared to the other 44 months of each 4-year election cycle.

Recall the basic formula: stock market value = future corporate profits / uncertainty. The stock market moves higher when one or both of two things happens: the expectation for future corporate profits increases and/or uncertainty decreases. What happens as an election nears? Outcomes become clearer – obviously not 100% clear – but clearer than several months earlier when so much was uncertain. As uncertainty wanes, stocks move higher and investors get compensated for the uncertainty. So, while the natural inclination may be to avoid uncertainty ahead of an election, the economic rewards transfer to those who are willing to assume the uncertainty of investing in advance of known outcomes.   

Fiction 2: Once the election result is known, you should reduce/increase risk if Trump/Biden/Elvis is elected.

Here is an important reminder of how investment markets work: once election results are known, stock and bond prices immediately reflect the result, positive or negative. The market immediately incorporates the future impact of anticipated tax policy and fiscal spending and moves accordingly. Stocks within politically-sensitive sectors like healthcare, defense, and energy instantly adjust up or down based on the election result. There is no free lunch where investors can load up on defense contractor stocks after the elections results are known and expect to be rewarded because Republicans swept the ballot. Any such news will already be factored into stock prices by the time that any of us can trade on the results.  

Unless the election result is a big surprise, the market is likely to have already “baked-in” the anticipated policy of the frontrunner candidate(s) before election day. As a result, the rational justification for trading based on election results is that you can better forecast the impact on global economics than the best institutional investors in the world. Citing investor Allan Roth in the Wall Street Journal, “Extrapolating common knowledge into predictions of the future is really just following the herd, and that typically doesn’t go well.”

Fiction 3: The election result will significantly impact on the stock market over the subsequent four years.

For months leading up to election day, we are inundated with politics, pundits, polls, and partisanship. It saturates the headlines and stories we read, hear, and see. The natural conclusion is that the election result will have a profound impact on the stock market. However, the impact of the election is likely far less impactful on stock and bond markets than we believe. The obsession to so closely link politics and Wall Street is radically misplaced. 

In fact, we make the same mistake of misjudging the impact of a CEO on a business’s performance or a coach on a team’s success. While the CEO, the coach, or the president do have an impact, all the evidence shows that we tend to dramatically overestimate their importance.

What has far more impact on the stock market than the president? Start with Federal Reserve policy. Anyone who is still unclear on why the stock market advanced more than 55% from its March 2020 lows despite a global pandemic and the deleterious economic consequences is likely underselling the impact of 2.3 trillion dollars being added to the economy in April by the Federal Reserve and the follow-up monetary stimulus. Historical data clearly shows that Fed policy has a far more direct and significant impact on the stock market than who sits in the Oval Office.

An additional factor that is considerably more predictive of future investment returns is the value of the stock market on election day. Said differently – the cheapness or richness of stocks on election day offers a far better indication of how the market will behave over the ensuing four years than who wins the election. 

For some validation, consider the S&P 500 cyclically adjusted price/earnings (CAPE) ratio – a useful measure of the stock market’s valuation. We can bifurcate each 4-year presidential cycle based on whether the starting CAPE on election day was above or below the median CAPE. On the left side of the charts below are the presidential terms that began when the stock market was cheap (relative to the median) and on the right side are the presidential terms that began when the stock market was expensive.  

The average annual return for presidents lucky enough to begin their term with cheap stock market valuations (left side): 10.2%. The average for unlucky presidents, elected when stocks were at more expensive valuations (right side): 0.8%1. 

Fiction 4: Republican presidents are better for the stock market.

The data is very clear to the contrary. Since 1933, the S&P 500 has achieved a 10.2% annual real return with Democrats in the White House compared to a 6.9% real return when Republicans occupy the White House2. There are, however, many problems with taking this data and drawing conclusions from it as we will explain further below.

Fiction 5: Democratic presidents are better for the stock market.

Democrats will point to historical data and conclude that stock markets and corporate growth perform better when a Democrat is in the White House. The reality is that this deduction confuses causation and correlation. There is no way to perform a scientific experiment where the economy goes into a laboratory to test for the controlled impact of one element (presidential party). Way too much noise. Outside factors such as changes in oil prices, global monetary policy, terrorist events, natural disasters, asset bubbles, and technological advancements have far more impact on stock prices than presidential policy initiatives. We cannot just strip out the impact of these exogenous factors and view the impact of the presidency. 

Furthermore, the data set is still too limited to draw significant conclusions given that there have been only 8 Republican and 7 Democratic presidents since the inception of the S&P 500 Index. Further to this point, nearly all of the performance advantage for the Democrats can be attributed to the stock market boom under Bill Clinton and the subsequent dot.com meltdown under George W. Bush.  

Fiction 6: A Biden presidency will result in higher corporate taxes and, consequently, be negative for the stock market.

Rather than focusing on the historical stock market record of past presidents, investors are right to be focused on the policy agenda of each candidate and the resulting investment ramifications. Candidate Joe Biden has outlined a plan to reverse the Trump tax cuts for corporations which, if implemented, would hurt corporate profits. Wall Street economists estimate that the proposed corporate tax hikes would reduce 2021 corporate profits by 5-12%.

While accurate in a vacuum, this logic ignores two key factors. First, Biden would need Congressional help to get his proposed tax hikes passed – something that seems unlikely if Republicans retain Senate control. Second, this logic considers only one component of Biden’s policy initiatives. Just as Trump’s foreign trade war had a negative impact on corporate profits that countered the beneficial tax cuts, Biden’s proposed tariff reversals would similarly serve as an offset to corporate tax hikes in regards corporate profits. The same could be said for Biden’s proposed $2 trillion infrastructure spending package which would be kindly welcomed by the stock market (albeit not kindly welcomed by the bond market).

What it all means to you

Here is how November 2020 will play out: On election day or soon thereafter, we will learn the election results. Roughly half the country will be disappointed – if not outright distraught – by these results. Many individuals in this disappointed/distraught camp will look to their investment portfolio as a way to respond. They will make dramatic changes, rationalizing such actions based on steadfast concerns about the direction of the domestic economy, tax policy, foreign trade, government spending, and the leadership of our country. Although the reasoning may seem perfectly rational, these will be emotional, personally biased, knee-jerk responses.

There is clear evidence that our political beliefs and the political climate impact our investment behavior. Citing a comprehensive examination of politics and investment decisions, “Individuals become more optimistic and perceive the markets to be less risky and more undervalued when their own party is in power. These shifts in perceptions of risk and reward affect investors’ portfolio decisions.” Furthermore, the historical data demonstrates that investors “improve their raw portfolio performance when their own party is in power.” This performance advantage comes not because one party is better than the other for stock markets. It comes because investors who stay invested achieve better returns. When investors allow their disappointment with political results to impact investment decisions, the implication is worse investment returns – a double whammy of sorts.   

The best course of action after election day will be to fight back any emotional responses – good or bad – and adhere to the same pre-election long-term discipline. We recognize there will be disappointment with the election result, regardless of who wins, but we are confident that the strict discipline of removing emotion and knee-jerk reactions from investing decisions will prove better in the long run.