The Tax Cuts and Jobs Act’s Impact on Roth IRA Conversion Planning

One of the greatest retirement account benefits in the entire tax code was dealt a death blow by the Tax Cuts and Jobs Act (TCJA). Prior to the law’s enactment, clients had the ability to recharacterize (undo) a Roth IRA conversion up until October 15th of the year following the year the conversion was made. This allowed for a host of benefits and planning strategies that financial planners were able to exploit for the better part of 20 years. Unfortunately, the Tax Cuts and Jobs act eliminated the recharacterization for Roth IRA conversions made in 2018 or later.

In the initial weeks after the passing of the Tax Cuts and Jobs Act, there was some confusion over whether or not 2017 Roth IRA conversions would be able to be recharacterized. The Tax Cuts and Jobs Act’s language called for the prohibition  on such transactions to apply to “taxable years beginning after December 31, 2017.”  Did that mean that only conversions made on or after January 1, 2018, could no longer be recharacterized, or that no recharacterizations could be executed on or after January 1, 2018, at all?

Experts were initially divided, but thankfully, we didn’t have to guess for long. In January 2018, the IRS released guidance to clarify that the ban on recharacterizations of Roth IRA conversions only applies to conversions made on or after January 1, 2018. Thus, if you have clients who made Roth conversions in 2017, they still have until October 15, 2018, to change their minds and recharacterize all or a portion of those conversions.

It remains to be seen though, exactly how helpful this new guidance will be for advisors and their clients. Given the uncertainty of recharacterizations of 2017 Roth conversions until early 2018, out of an abundance of caution, many advisers recommended their clients consider making recharacterizations of any 2017 Roth conversions they weren’t sure they wanted to keep before the end of last year (2017).

Note: The elimination of recharacterizations only applies to Roth IRA conversions, not Roth IRA contributions. Thus, if, for example, your client makes 2018 a Roth IRA contribution today, but finds out they are over the 2018 Roth IRA contribution income limit when they prepare their 2018 return next year, they can still recharacterize that Roth IRA contribution to a traditional IRA contribution.

Nevertheless, for those clients who made 2017 Roth conversions and have at least some portion of that conversion remaining (that wasn’t recharacterized), it’s good to know that the flexibility to undo the balance of the transaction still exists until October 15, 2018. Common reasons to consider recharacterizing a 2017 Roth IRA conversion by the October 15, 2018, deadline include:

  • The client’s account dropped in value since the conversion – If your client made a Roth IRA conversion, but the amount converted drops in value by October 15, 2018, a recharacterization may make sense. Why pay tax on value that no longer exists if there’s no requirement to do so?
  • The tax bill is more than they expected – Sometimes clients underestimate their income for a year or overestimate their deductions. Then again, those items are often difficult to predict. Maybe your client’s company had had a banner year and your client received a sizable bonus at the end of 2017 that wasn’t expected. That might have pushed them into a higher tax bracket and made the conversion less valuable. If that’s the case, the client can simply recharacterize all or a portion of their conversion to mitigate the tax bill. Remember though, that same flexibility won’t exist going forward.
  • Your client changed their mind – Maybe the Tax Cuts and Jobs Act’s changes mean that your client will pay a lower tax rate in 2018 and for the foreseeable future than had previously been anticipated. If so, maybe it’s best to recharacterize the 2017 Roth IRA contribution and make it in 2018 instead. That said, there’s nothing stipulating the precise reasons a Roth IRA recharacterization can be made. So regardless of your client’s reasoning, if they want to undo their 2017 Roth IRA conversion, they can do so by October 15, 2018, and enjoy this last-of-its-kind opportunity.

For all new Roth IRA conversions, it’s extraordinarily important for advisors to understand (and to make sure their clients understand) that, as a result of the TCJA, clients are now “stuck,” for better or for worse, with each Roth IRA conversion they make, along with the resulting tax bill. That makes doing the homework before making a Roth IRA conversion more important than ever before.

In the past, the ability to recharacterize made the Roth IRA conversion recommendation fairly benign. Outside of some extra paperwork, a “bad” conversion wouldn’t really hurt a client all that much so long as it was timely recharacterized. In a sense, the old rules made it less a question of “how much to convert” and more a question of “how much to recharacterize.”

Today, however, planners no longer have the luxury of a mulligan on the Roth IRA conversion recommendation. Therefore, you need to do a lot more heavy lifting on the front-end of the transaction. At the very least, before suggesting a Roth IRA conversion, you should typically have a:

  • Rough estimate of the conversion’s potential impact on your client’s tax bill – In previous years, it was helpful to have an idea of how a Roth IRA conversion would impact a client’s tax bill before the conversion was made, but it wasn’t absolutely essential. For example, suppose a client was trying to stay within their 15% (pre-TCJA) tax bracket and made a $30,000 conversion. Now imagine that, upon completing their tax return for the year, the client discovered that they were actually $10,000 over the 15% bracket. Under the “old” rules, this wasn’t a big deal, as the client could simply recharacterize $10,000 of their initial conversion and back into the exact amount of taxable income they wanted to show on their tax return. It was almost too good to be true (which is no doubt why Congress took aim at the provision in the first place). If that same conversion were made today, however, the client would be unable to reverse course in the same manner.

Clearly, having a solid understanding of what a client’s Roth conversion will do to their tax bill (and Medicare Part B/D premiums, etc.) prior to completing the transaction is more important than ever before. The problem, however, is that for many clients, it’s difficult, if not next to impossible to make an accurate prediction of income and/or deductions early in the tax year. This may very well lead to a major change in the paradigm of Roth IRA conversion planning (as discussed below in A Change in the Optimal Time During the Year to Convert?).

  • Reasonable expectation that the benefits of “pre-paying” taxes via a Roth IRA conversion make sense in the client’s overall plan – The number one driver of the to-convert-or-not-to-convert decision is to determine when a client will be able to pay tax at the lowest rates. If today’s rate would be lower than the future rate the client would otherwise pay, then a Roth IRA conversion is often advisable. Conversely, if a client is confident that their future tax rate will be lower than the rate they would pay today on a conversion, such a conversion may not make sense.

The problem with this, of course, is that the future tax rate a client will pay is an unknown. The best we can do today is to make reasonable projections. Ultimately, that future tax rate will be a function of two factors; the amount of income a client has which is subject to income tax, and the tax rates in effect at the time. A change in either of these variables can make a big difference in whether or not a Roth IRA conversion today proves to be a good financial decision. Due to these future unknowns, some clients may prefer to have at least some money inside a Roth account as a hedge against a rising tax rate. This “tax insurance” can be the way a conversion makes sense as a part of their overall financial plan, as long as it’s understood by the client.

And thanks to today’s “low” tax rates, for many clients, the cost to purchase that tax insurance is as affordable as ever. While the Tax Cuts and Jobs Act’s changes won’t be good for everyone, most Americans are expected to see lower tax bills (through 2025, after which the majority of TCJA changes to individual income taxes will revert to the “old” rules). Furthermore, most estimates show that the majority of those benefits, expressed both in terms of dollars and as an overall percentage, will be skewed toward the higher income clients many advisors target. As such, even without the ability to recharacterize, a Roth IRA conversion may make sense for more clients today than ever before.

  • Strategy for paying the additional taxes your client will owe as a result of the conversion – In the past, if a client underestimated the tax bill due from a Roth IRA conversion, or experienced a financial hardship shortly after their conversion and was no longer comfortable footing that bill, they were able to use the recharacterization to undo the transaction and wipe away all or a portion of the tax due. With that no longer being an option, the Latin expression “caveat emptor,” or buyer beware, must apply. Clients must have a solid, tax-efficient plan to pay for the conversion. Typically, the tax bill should not be paid with other IRA funds fund, nor should it be paid with newly converted Roth IRA assets. If assets in a taxable account are to be sold to help pay help pay the tax bill, any additional tax owed as a result of those liquidations needs to be factored in as well.

A Change in the Optimal Time During the Year to Convert?

In years prior to 2018, a  general best practice for Roth IRA conversions was to make them as early in the year as possible. One reason for this logic is that, in general, markets tend to rise over time (though there are certainly many years in which market returns are negative). Converting earlier in the year, therefore, would allow the growth on the converted assets to occur inside the Roth IRA, as opposed to inside the traditional IRA.

For example, take a client making a $100,000 Roth IRA conversion in 2017. 2017 was a pretty great year for markets across the board. If that client had made their Roth IRA conversion in January and used an aggressive asset allocation, they might have seen $20,000 (20%) of presumably tax-free growth inside their Roth IRA during 2017. If, on the other hand, they had made their Roth IRA conversion on December 31, 2018 – just to take it to the extreme – that same $20,000 of earnings would have occurred inside a traditional IRA, and would ultimately be taxable when distributed. As you can imagine, for those making regular conversions over a number of years, converting earlier in the year as opposed to late in the year could have a major impact on long-term tax efficiency.

Another reason that converting earlier in the year was often the better option was that it gave you the most amount of time to evaluate whether making a recharacterization of that conversion was advisable. Similar to the time value component of an option’s price, in a sense, the recharacterization “option” was worth more the earlier in the year a conversion was made. A Roth IRA conversion on January 1st of a year, for instance, gave you 21 ½ months in which to evaluate portfolio performance, the precise tax impact of the conversion, and other factors that played into the keep-the-conversion-or-recharacterize decision. Converting at the end of the year gave you less than half that time to make the same analyses.

While a longer evaluation period was preferable, the fact that, in a worst-case scenario, a client still had 9 ½ months during which to change their mind meant that any surprises in income were of minimal concern. Big year-end bonus? No problem! Deductions lower than expected? No big deal. An inefficient conversion could always be timely reversed.

Now, however, there’s an argument to be made that the paradigm should shift towards conversions made later in the year when a client’s taxable income is able to be more accurately projected. Admittedly, this is a much harder task for some clients than it is for others. For instance, a client living off of their pension and Social Security benefits, taking only their required minimum distributions, and using the standard deduction will have fairly predictable taxable income from one year to the next.

On the other hand, other clients may have taxable income that is much less stable from one year to the next. Business-owner clients, for instance, often have taxable income that fluctuates significantly from one year to the next. This could be due to a down year in business or a particularly strong one. It may also be the result of non-regular expenses. For instance, suppose your client needs to upgrade a major piece of machinery every four to seven years. If that machinery cost $400,000 and the entire amount is expensed in the year of purchase via bonus depreciation, then it stands to reason that every four to seven years, your client may have a year in which a Roth IRA conversion should be given extra consideration.

Your client may have some idea as to when the next replacement will occur, but the precise timing may be somewhat up in the air. Will it be a fourth quarter 2018 purchase, or a first quarter 2019 purchase? That answer may not be known at the start of the year, and the difference of a few weeks in the timing of the purchase could have significant tax implications. A Roth IRA conversion could turn out to be a great move, or a terrible one.  As such, for clients like this with less predictable income, it may be best to wait until later in the year, when income and deductions are easier to project, to make any Roth IRA conversions.

Clearly, there are advantages to making early year Roth IRA conversions, but there is also more of a reason to wait to make those conversions today than ever before, so what’s an advisor to do to get the best of both worlds? Here are two strategies to consider:

2018 Roth IRA Conversion-Cost-Averaging

Instead of making a single conversion late in the year, you could, instead, make multiple Roth conversions throughout the year. This is not exactly a new idea, but in practice, it will work much differently than In the past.

For instance, suppose that in 2016 your client wanted to convert $50,000 to a Roth IRA, and in early January of 2016, they executed a “full” $50,000 conversion. You may recall that 2016 began with one of the worst early-year market declines in history. As such, if your client had another $50,000 in an IRA (or another eligible account), they could have executed another “full” $50,000 Roth conversion after the drop to take advantage of the lower valuations with the hope that any recovery would occur inside the Roth IRA. If the market went up from there, then you could simply recharacterize the first “full” conversion and leave only the second conversion intact. On the other hand, if the markets had continued to go down and if the client had yet another $50,000 in their traditional IRA, they could have executed a third “full” conversion of $50,000 and recharacterized the first two conversions… and so on. At some point (hopefully), one of the conversions would be at or close to the bottom.

The elimination of the Roth IRA conversion recharacterization renders that approach null and void. Clients can still execute multiple conversions within the same year, but none of them will be able to be undone. Thus, Roth IRA conversion-cost-averaging needs to adjust in light of the new rules. Instead of making several “full” conversions and keeping only the best conversion (while recharacterizing the others) Roth IRA conversion-cost-averaging will now consist of several “partial” conversions made throughout the year.

For example, suppose that an initial projection of 2018 income leads you to believe that your client should convert $60,000 to a Roth IRA during the year. Instead of making one big conversion in January (in which case you risk over-converting  because you underestimated the client’s other income for the year or overestimated their deductions) or one big conversion in December (in which case you might have a strong handle on their taxable income for the year but may have already missed out on a full year’s worth of appreciation), you could make a $5,000 conversion each month. Utilizing this approach provides a balance of sorts, and has the potential to offer the following benefits:

  • If the market appreciates over the course of the year, at least a portion of that appreciation will be earned within the Roth IRA., (though not as much as if the entire conversion had been made at the start of the year).
  • If the market declines over the course of the year, the client will be making at least a portion of the year’s conversion total at lower valuations (though not as much as if the full conversion took place after the drop).
  • You can make “course corrections” on the target conversion amount throughout the year as the client’s tax picture becomes clearer. If, for instance, as the year progresses it appears that the client is having a better year income-wise than had initially been anticipated, you could cut your monthly conversions back to $4,000 a month. Alternatively, you might just skip a few months entirely towards the end of the year. And of course, if a bigger conversion than was initially planned for becomes the optimal tax play, you could easily increase the monthly conversion amounts. Another option, of course, would be to make the last conversion of the year a little bigger to make up any difference.

Operationally, this may seem rather onerous, and for some clients that may be the case, but for many clients, this strategy would not be particularly difficult to implement and monitor. Many custodians, for instance, will allow you to set up ongoing, monthly conversions with the submission of a single form. And as far as the tax reporting goes, it makes no difference whether a client makes twelve $5,000 Roth IRA conversions throughout the year, or a single $60,000 conversion. There will be one 1099-R distributed early the following year that the client will need to report on their tax return. Furthermore, if the client is younger than 59 ½ and needs to keep track of each of their conversions (for purposes of determining if potential future pre-59 ½ distributions of converted amounts from the Roth IRA are subject to the 10% early distribution penalty), all of the conversions will be considered a single conversion, made on January 1, 2018.

One final note on this strategy… no one knows for sure when the next big market correction will come. Many believe that there’s room for the market to run for another few years or longer. Others believe that we’re long overdue for a major correction. In the event the latter proves to be accurate, a Roth conversion-cost-averaging approach could prove to be an especially useful tool.

Summary of Roth Conversion-Cost-Averaging Strategy

Step 1: Project your client’s taxable income for the year.

Step 2: Determine how much it would make sense to convert if your client’s income is exactly as projected. If the answer here is $0, stop. Do not initiate any conversions yet.

Step 3: Divide the amount determined in step #2 into monthly, bimonthly, quarterly or other regular interval conversions.

Step 4: Begin making regular Roth IRA conversions as determined in step #3.

Step 5: Throughout the year, regularly update your taxable income projections and adjust remaining conversions accordingly.

Roth IRA Conversion “Barbelling”

The Roth IRA conversion-cost-averaging strategy may be a little much for some clients and advisors who wish to keep things simpler. Making quarterly conversions instead of monthly conversions could help simplify things and would likely be sufficient for most clients. Alternatively, advisors could engage utilize another approach we’ll call Roth IRA conversion “barbelling.”

If you’re familiar with the strategy of creating a bond barbell, then you might already know where this is going. If you’re not familiar with a bond barbell, here’s a super-high-level overview… a bond barbell simply describes a strategy in which a bond portfolio is split into two tranches. The first tranche is generally invested in short-term bonds to provide liquidity (for spending or, more often, to reinvest if interest rates are expected to rise in the short-term). The second tranche is generally invested in longer-term bonds to take advantage of the higher interest rate that is generally associated with longer duration bonds.

The Roth IRA conversion equivalent for this approach calls for advisors to plan for up to two conversions per year; one conversion, made as early in the year as possible, and a second conversion, made later in the year when the client’s tax picture is clearer. Again, the idea here is to get as much appreciation to occur within the Roth IRA shell as possible without “over-converting” and leaving a client with an inefficient transaction.

So here’s the question, how to you “know” how much to convert early in the year vs. later in the year? While there’s probably more than one reasonable answer to that question, one logical approach would be to convert as much as you are “sure” you’ll want to convert early in the year.

For example, suppose your client, Jenny, is a sales rep for a local auto supply store. She’s compensated by a combination of base salary and commission. Due to the commission portion of her compensation, Jenny’s earnings tend to fluctuate between $110,000 and $130,000 annually. Now let’s suppose that if Jenny has a slower year and only earns $110,000, it would be efficient to convert $50,000. Similarly, if Jenny has a stronger year and earns $130,000, it would only make sense to convert $30,000.

In a situation like this, it might be best to convert $30,000 earlier in the year, since that’s the minimum amount it would make sense to convert in a reasonably foreseeable scenario. Then, as the year progresses and Jenny’s income becomes more predictable, you could determine how much to convert as part of the second, late-year conversion. The higher Jenny’s earnings, the smaller that second conversion.

Summary of Roth Conversion “Barbelling” Strategy

Step 1: Determine a reasonable range for the client’s taxable income for the year.

Step 2: Determine how much it would make sense to convert if the client’s income was at the low end of the range. If the answer here is $0, stop. A Roth IRA conversion likely does not make sense for your client this year. 

Step 3: Convert the amount determined in step #2 as early in the year as possible.

Step 4: Towards the end of the year, when taxable income can more accurately be forecast, project the client’s taxable income again.

Step 5: Using the results of your analysis from step #4, determine the additional amount that can be tax efficiently converted for the year and complete that conversion before the end of the year (if the client’s funds eave the distributing account by December 31, 2018, it will be considered a 2018 conversion, even if the funds don’t make it into the Roth IRA until 2019).

Ultimately, whether or not a Roth IRA conversion is right for your client, and the strategy you choose to employ to execute such a conversion will depend on a client’s specific set of facts and circumstances. It will be important to take into consideration both quantitative and qualitative factors and to be sure clients are aware of the permanency of the transaction. There’s no doubt that that the Tax Cuts and Jobs Act has added some new complications to an already complicated decision-making process, but the good news for advisors is that their guidance and counsel may be more needed today than ever before.

Share, Email, or Print This