Archives for March 2021

Timing the stock market is hard. It’s so hard, in fact, that there’s a reasonably broad consensus that it’s unwise to even try.

And that’s in spite of the fact that, when attempting to time the stock market, you’re really only trying to predict one thing: what will the market do next? And there are only three possible outcomes: go up, stay flat, or go down.

Trying to predict specific changes to our tax law in any way that is useful for tax planning is far, far harder than trying to time the stock market.

Take one example: will distributions from Roth IRAs be made taxable in the future? And how should a person adjust their tax planning based on that possibility?

People ask me about that (and other similar topics) all the time. And it doesn’t seem like it should be that complicated — either some law gets passed that makes Roth IRA distributions taxable or no such law gets passed. Two possible options, right?

Unfortunately, no. There’s also the possibility that it would work like taxation of Social Security benefits. That is, we could see a change where Roth distributions remain nontaxable, unless your income is over a certain threshold. And this third possibility is itself an unlimited number of possibilities, because the threshold could be set at any level. And the calculation (for what percent of your Roth distributions are taxable, based on the amount by which your income exceeds the chosen threshold) could work in any number of different ways.

In other words, you could make a prediction that a bill will be passed that causes Roth distributions to be taxable. And you could be right about that prediction, but you have to get the details right too. If you get the details wrong, any tax planning decisions that you make based on your prediction are as likely to be harmful as helpful.

And if you do somehow manage to get the details right?

Well, there’s still the issue of timing.

You might precisely predict what this new provision in our tax code looks like — but be off by ten years as to when the legislation gets passed. And if that is the case, then you’re again in a situation where any tax planning decisions that you make based on your prediction are as likely to be harmful as helpful.

And even if you are right as to the details of how the new provision works and you are right as to the timing, there’s yet another problem: tax legislation never changes just one thing. Most pieces of tax legislation are hundreds of pages long.

Even if you accurately predict the details of the provision in question and when it will be implemented, some other change made by the same legislation could completely mess up the math in your analysis. And once again you’re left in a situation where any tax planning decisions that you make based on your prediction are as likely to be harmful as helpful.

A vague sort of prediction that “marginal tax rates for people with high incomes might be higher later than they are now” is pretty reasonable. But trying to guess at the specifics of any particular change — and then make specific tax planning decisions accordingly — is just not worth the time and effort.

A reader writes in, asking:

“I wonder if you can write about the ‘marginal tax rate is not necessarily the same thing as your tax bracket’ concept for people who are still working. My spouse and I got caught in this situation for the first time in the tax year 2020 because we will be subjected to the ‘Investment income tax’. We will be addressing this by changing part of our Roth 403b contribution to regular 403b contribution for 2021 and beyond (i.e. to keep our MAGI as a married couple to < $250,000).”

As we’ve discussed many times in prior articles, your marginal tax rate is often higher than just your tax bracket. Most often, I’m discussing that concept in the context of retirees, due to the way Social Security benefits are taxed and the way Medicare premiums are determined. But it can apply to people in their working years as well.

Things that can cause such an effect for people still working would include:

But here’s where things get tricky. With total taxable income of $42,000, Bob is in the 22% tax bracket. But if his ordinary income increases by $1,000, his income tax for the year will increase by $270 (i.e., a 27% marginal tax rate).

That’s because the 12% bracket extends up to taxable income of $40,525. And, excluding LTCGs, Bob’s taxable income is below that threshold. However, his taxable income excluding LTCGs is now $39,000 rather than $38,000. So we have 12% regular income tax on this $1,000 of additional income (i.e., $120 of tax).

But there’s another effect going on as well. There is now only $1,400 of space for LTCGs before hitting the $40,400 threshold. That is, this $1,000 of additional ordinary income is also causing another $1,000 of his long-term capital gains to be above the $40,400 threshold and therefore taxed at a 15% rate, so that causes another $150 of income tax.

So we have $120 + $150 = $270 total dollars of tax being caused by this $1,000 of income. A 27% marginal tax rate, despite a taxable income that is in the 22% bracket.

After the recent article about maximizing after-tax dollars (as opposed to minimizing taxes), several people wrote in to ask about Roth conversions — specifically, when does it make sense to do one?

The answer depends significantly on how you are going to pay the tax on the conversion: would you be using money from the traditional IRA, or would you be using money from a taxable account? For example, compare the two following scenarios, each of which involves a conversion at a 20% anticipated marginal tax rate.

Example #1: Charlie takes $50,000 out of his traditional IRA and moves $40,000 of it to a Roth IRA. The remaining $10,000 will be used to pay the tax on the conversion.

Example #2: Kiara takes $50,000 out of her traditional IRA and moves all $50,000 of it to a Roth IRA. She will use $10,000 from her regular checking account to pay the tax on the conversion.

Using Retirement Account Dollars to Pay the Tax

For cases in which dollars from the traditional IRA would be used to pay the tax on the conversion, it’s purely a question of marginal tax rate.* That is, each year, for each dollar in the traditional IRA, you would ask what your current marginal tax rate would be if you converted that dollar right now, and you compare that to what the marginal tax rate would be for that dollar if you did not convert (i.e., what tax rate would be paid whenever the money comes out of the account later).

If the current marginal tax rate is lower, then a conversion is advantageous. If the current marginal tax rate is higher, then a conversion would be disadvantageous.

To back up a step, a traditional IRAs can be roughly thought of as a Roth IRA, of which the government owns a portion. For example, if you expect a 25% marginal tax rate in retirement, a traditional IRA is much like a Roth IRA, of which you own 75% and the government owns 25%.

When you do a Roth conversion, you’re essentially “buying out” the government’s share of the converted dollars. If your current marginal tax rate is lower than the marginal tax rate you expect later on (i.e., whenever you would be distributing the dollars in question), then you’re buying out the government’s share at a bargain price (e.g., paying 15% now when you would otherwise be paying 25% later). If your current marginal tax rate is higher than the marginal tax rate you expect later on, then you’re buying out the government’s share at a high price (e.g., paying 30% now when you could instead pay 25% later).

Using Taxable Dollars to Pay the Tax

If, however, dollars from taxable accounts (e.g., just a regular checking or savings account) would be used to pay the tax on the conversion, then the analysis changes somewhat. You’re using non-retirement-account dollars to buy the government’s share of your IRA. And that, in itself, provides some value.

The key point here is that taxable accounts grow at a slower rate than IRAs, because you have to pay tax on interest/dividends each year. So now there are two things going on with the conversion:

1. As in the prior case, you’re buying the government’s share of the IRA now, rather than later (i.e., paying at your marginal tax rate now rather than your future marginal tax rate), which could be advantageous or disadvantageous, and
2. You are using taxable account dollars to buy IRA dollars, which is advantageous.

In the analysis, marginal tax rates are still super important (because of point #1).

But now we’re also concerned with time frame and rates of return (because of point #2). Again, dollars in an IRA grow at a faster rate than dollars in a taxable account, because you don’t have to pay tax each year on the interest/dividends. The greater the length of time that these dollars will remain in the IRA, the more impactful that fact becomes. That is, the money will be compounding at a faster rate in the IRA, but if that’s only happening for a few years, that’s not so important. If it will be happening for a few decades, it’s super important. And expected rate of return matters as well. If the expected rate of return (before considering taxes) isn’t that high to begin with, then the tax cost within a taxable account isn’t so great. Conversely if the expected rate of return is very high, then the cost of having the money in a taxable account becomes much more significant.

The result of all of this is that, if you’re using taxable dollars to pay the tax, then, depending on time frame and expected rate of return, it might even be advantageous to do a Roth conversion if the current marginal tax rate is higher than you would expect it to be in the future.

*I know I harp on this over and over, but it’s critical to understand that your marginal tax rate is not necessarily the same thing as your tax bracket. In many cases, especially in retirement, your marginal tax rate will be greater than your tax bracket because additional income not only causes the normal amount of income tax, it also causes something else undesirable to happen (e.g., it causes a particular credit to phase out, it causes more of your Social Security to become taxable, or it causes your Medicare premiums to increase).