There is a refreshingly honest sentence at the top of this weekend’s Tax Report: the great obstacle to effective tax planning for retirement funds is Congress. To make smart moves, savers need to know about future taxes — and this is impossible. Will lawmakers tackle $40 trillion of debt by raising rates, altering provisions, or adding a consumption tax? Maybe. Maybe not. Then, on top of that, you have to predict your own future: how long you will work, how much you will save, whether income in retirement rises or falls.
The upshot, as the column puts it, is that tax planning for retirement accounts comes down to guesses. A previous column on this subject drew hundreds of reader responses. Some derided affluent savers for complaining about good fortune — one wrote, in effect, that the steak was too juicy and the lobster too buttery. More were grateful. One said simply: I wish I had read this 20 years ago.
The framework, and the number that decides it
Edward McQuarrie, a professor emeritus at Santa Clara University who studies retirement strategies, offers the sensible version: tax planning involves guesses, but it is important to make smart guesses, because a lot of money is at stake. The defense against not knowing is diversification — across traditional accounts, Roth accounts and, in some cases, taxable brokerage accounts. Ideally a saver puts after-tax dollars into Roth accounts when their rate is low and takes deductions for traditional contributions in higher-earning years. Always take the employer match regardless.
But that ideal was not available to most mid- and late-career savers, who had no Roth option when they were young. For them McQuarrie warns against compounding one problem with another by overpaying for conversions. His example is clean: if John’s top rate is 22% now and would be 22% on withdrawal, and a conversion is large enough to be taxed at 32%, he should think about skipping it unless he is convinced rates will go way up. If instead John retires into a few low-income years in the 12% bracket before required withdrawals begin at 73 or 75 depending on birth year, conversions could be smart. The rate you convert at versus the rate you would have withdrawn at is the whole calculation.
Four situations that trip people up
On required withdrawals: yes, you can still convert, but the required amount cannot be part of the conversion and has to come out first — which raises the cost. On a market crash after converting: a traditional account effectively shares its losses with the government, because a smaller balance means smaller tax on withdrawals. A Roth does not share, since the tax was already paid. Ed Slott, a CPA and IRA specialist, puts it as: always try to leave funds in Roth accounts as long as possible.
On charitable giving, the answer is close to unambiguous. In 2026, owners at least 70½ can make Qualified Charitable Distributions of up to $111,000 directly to charities. They count against required withdrawals and do not raise adjusted gross income, which is what triggers other taxes downstream. For anyone charitably inclined with a large traditional account, that is among the most efficient moves available. And on long-term care, a caution that catches people: care costs are often deductible above 7.5% of adjusted gross income, so converting away too much of the traditional balance can leave too few taxable dollars available to generate those deductions.
On the widow’s penalty — the higher rates a surviving spouse faces after switching to single filing status — the specialists genuinely disagree. Slott thinks it is often a good reason to convert, especially on a final joint return. McQuarrie thinks in most cases it is not. That disagreement is not a failure of the article. It is an accurate picture of the field, and anyone who tells you this question has one clean answer is selling something. What is clear is that if there are other reasons to convert, the widow’s penalty does not negate them.
None of this is advice for anyone in particular, and every item above turns on a specific bracket, a specific balance and a specific year. But the shape of the work is the same for everyone: know your current rate, make a defensible estimate of your withdrawal rate, and stop converting where those two cross. That is genuinely a fifteen-minute conversation if you bring the statement and last year’s return — and it is one of the few conversations where the answer is worth real money rather than reassurance.