A relative asked me about Roth conversions. I can never think of any good reasons to do one, and I especially don’t understand why investment advisers suggest them: Doesn’t it give them less money to play with and therefore, less commission income? I will admit that if someone is moving from Nevada to California, or Florida to New York, it might be a good idea to avoid state taxes. But most of the traffic goes in the other direction.

It seems like a simple formula to me: If you have $1,000 and want to pay $200 tax on it now, you have $800 left to invest. Let’s say you can earn 10% for the next five years. You end up with about $1,289 tax-free. Or, you can keep your $1,000 and invest it at 10% for five years. You’ll have about $1,611 and, after paying tax on it at 20%, you’re left with about $1,289.

So it might be a good idea if you expect your tax bracket to increase, but that involves predicting what Congress is going to do with tax rates every year or so. And I have seen clients withdraw thousands from their IRAs, tax-free, to pay assisted-living and other medical expenses in their final years. The best bracket is zero percent, even if you arrive there with unfortunate deductions.

Back before the last recession, I saw a client pay about $30,000 in tax to convert $100,000 to a Roth – he invested it in tech stocks, which quickly lost more than half their value.

Am I missing something here?