Like so many topics in financial planning, the Roth conversion decision tends to be reduced to a set of assumptions and some financial modeling. But from my perspective, the math fails to address all the relevant aspects. We’ll get to what I think is left out but first, let’s start with some background.
Roth IRA accounts and Traditional IRA accounts have different tax features. With a Traditional IRA, an investor typically makes pre-tax contributions (you get a tax deduction upon contributing), the growth within the account is tax-deferred, and the investor later pays taxes when withdrawing from the account. With a Roth IRA account, an investor does not receive a tax benefit for making contributions (i.e., contributions are after-tax), but growth and any subsequent withdrawals are tax-free.
A Roth conversion describes a transfer from a Traditional IRA account (or simply, “an IRA Account”) to a Roth IRA account. When converting to a Roth IRA, the investor pays income tax in the year of the conversion. Converting dollars to a Roth IRA may be in the investor’s interest if the prevailing tax rate at conversion is lower than the tax rate that will later apply to their IRA distributions[1]. Here’s an example of how a Roth conversion can make good financial sense…
Molly and John are preparing to retire this year at age 67. Unlike in their working years, they will have relatively low taxable income for the duration of their 60s. Beginning at age 70, they anticipate receiving Social Security income. At age 73, they will take Required Minimum Distributions from their IRA accounts. The combination of those two income sources will cause their expected marginal tax rate for much of their 70s and beyond to be higher than in the first few years of retirement.
When they convert assets from IRAs to Roth IRAs, they anticipate paying taxes at a lower marginal tax rate than the tax rate they expect to have later in retirement. Of course, no one knows for sure what future tax rates will look like, but if they can convert at low enough tax rates in their 60s, they have a good chance for the math of this strategy to work in their favor.
Now I want to introduce some observations on investor behavior that I believe are relevant to the Roth conversion decision:
- Investors who are well-prepared financially for retirement are more likely to underspend their resources than risk overspending.
- Most individuals have a hierarchy for the psychological “ease” of spending from various income sources.
I would consider the first point to be an anecdotal observation if not for the published studies that document the so-called “consumption gap” phenomenon[2]. The consumption gap refers to how much retirees could reasonably spend from their portfolios versus what they actually spend. Studies confirm our observation that retirees spend less than they feasibly could. I think that for many of our clients, spending in retirement creates some anxiety. Unlike when they were working, they are no longer adding to their investments, and investment growth feels uncertain. Withdrawing from the pot of money that must last their lifetimes can induce a sense of loss that is similar to market volatility.
The second observation was first noted in the 1990s by a pair of behavioral economists[3]. My guess is that you have noticed this preference in your own life. Do you find it easier to spend your earned income rather than taking from savings or investments? It shouldn’t matter, but for some reason, it does. We create “mental accounts” and treat certain assets or streams of income differently.
So how do these observations relate to Roth conversions? They relate because of the mandatory distribution rules that apply to IRA accounts but do not apply to Roth IRA accounts. Under current laws, when you reach age 73, you must begin to withdraw money from your IRA accounts, with the minimum amount determined by an IRS formula. These required distributions force you to withdraw cash. My hypothesis is that the cash from IRA distributions is easier to spend for your lifestyle needs than if you otherwise needed to sell assets in other investment accounts. So, by taking money out of your IRA account and transferring it to a Roth IRA account (a Roth conversion), you are reducing the pot of money that will produce required distributions, which are easier to spend, and increasing a pot of money that you are less inclined to utilize!
Moreover, the tax-free nature of the Roth IRA encourages holding assets in that account for as long as possible. Therefore, Roth assets are often the last place a retiree will look for producing the cash they need to fund their lifestyle. I believe that most of our clients with Roth IRA assets will never touch the money in those accounts. If that’s the case for you, a Roth conversion amounts to pre-paying income taxes for your heirs, which is perfectly fine if that’s an objective of yours. But I’m not sure that is a primary objective for all who are considering Roth conversions.
My recommendation when evaluating Roth conversions is not only to consider the math (marginal tax rate now versus expected marginal tax rate in the future), but to also reflect on your key objectives and your likely retirement spending disposition (and that of your spouse, if married). If you think that you might have some anxiety with retirement spending when largely living from your investment portfolio, think about how much you want to reduce an account (your IRA) that will effectively nudge you to spend a bit more.
And if you’re already retired and have Roth IRA assets, here’s one more recommendation. If you don’t like the idea of leaving your Roth IRA untouched for life, pick something that you like doing each year, and fund it with a distribution from your Roth IRA. It’ll bring some purpose to your planning.
[1] Tax rate here should include Federal taxes, potentially state taxes, and other components such as the net investment income tax, IRMAA adjustments, etc.
[2] See, for instance, “Spending in Retirement: Determining the Consumption Gap”, Journal of Financial Planning, February 2016.
[3] Hersh Shefrin and Richard Thaler. For instance, see “Mental Accounting Matters” (Thaler), Journal of Behavioral Decision-Making, 1999. Interestingly, their real-world application suggestion was to increase the money in accounts that people feel are “off limits” to encourage aggregate wealth accumulation. My focus is the opposite!
