A reader writes in, asking:
“I do not believe in market timing and my approach of regular investing over the years without selling has been very successful. However, when you make a Roth conversion, within the year you are market timing.
From what I understand, I can tell my broker whatever date I desire to have my Roth conversion be effective. They will then send me a 1099 indicating the market value on the conversion date. The market value is then treated as taxable income on my federal and state returns.
However, within the year when I will be making my conversion, the market value of my retirement account could fluctuate materially. Should I do the conversion as early in the year as possible to give the longest time for tax free growth? Should I watch interest rates and make the conversion when rates come down? Or should I try to convert in portions throughout the year?
The additional tax caused by poor timing could be significant. I have not seen any discussion of this issue and I thought, in addition to me, it might be of interest to other readers.”
If the dollars that would be used to pay the tax would be coming out of the IRA itself, and the tax rate would be the same at different points during the year, then it doesn’t matter at all when during the year the conversion is done. Returns, whether positive or negative, as well as the payment of tax, are both multiplication functions. And the commutative property of multiplication tells us that we can do those multiplications in any order and end up with the same result. (That’s the rule from grade school math that tells us that A x B x C is the same as C x B x A.)
The tax rate might not be the same though, at different points during the year. For example, imagine that, on January 1, you have $50,000 in an IRA that consists of 1,000 shares of a particular mutual fund. If you convert the whole IRA on January 1, it would be $50,000 of income. Now imagine that, on March 1, those same 1,000 shares happen to be worth only $35,000. It’s possible that the actual tax rate would be different on an additional $50,000 of income rather than an additional $35,000 of income.
If the dollars that would be used to pay the tax on the conversion would be coming out of taxable accounts, then you’re effectively using taxable dollars to “buy more” Roth dollars. And the lower the share price on the date of conversion, the more effectively you are using your taxable dollars.
Overall point being, if you somehow knew in advance which day of the year would have the very lowest market value, that would be the best day to convert, because a) you might be able to pay a lower rate on the conversion (or a portion of the conversion), and b) if you’re using taxable dollars to pay the tax, those taxable dollars will go further, in terms of being able to pay for a larger number of shares being converted.
But, of course, there’s no reliable way to do that. No way to know when we’ll see the lowest market levels in a given year.
On average — because investments generally go up in value over time — earlier in the year will likely be a better “deal” than later in the year. Conceptually, this is the same as the question of lump-sum vs dollar cost averaging. That is, the earlier in the year you do it, the more growth you would be expected to take advantage of. Though whether it actually plays out that way in any particular year has a large chunk of randomness involved.
There is an important counterpoint though, with regard to conversions. Early in the year, many of the other numbers on your tax return may not yet be known. For instance, in January, you may find that it’s very hard to predict how much earned income, interest income, dividend income, or capital gain distributions from mutual funds you’ll have over the course of the year. Waiting until later in the year can make it much easier to try to target a specific income threshold with the conversion (e.g., staying just below a given IMRAA threshold or staying within a given tax bracket).
On the whole, I think for most people it makes most sense to wait until later in the year, given the uncertainty about all the tax planning inputs.
If, however, you find that you can pretty reliably predict most of the inputs on your tax return (e.g., you’re retired, so you know your earned income will be zero, and you find that your dividend income is pretty reliable each year), converting earlier in the year can make sense. Or, if at some point during the year you happen to notice a major market decline, you could go ahead and do a conversion at that time. (Though of course you’d have to accept the fact that you may be missing out on an even better opportunity that could arise a few days/weeks/months later. There’s just no way to know.)
Related reading: