The Effect of Taxes on the Social Security Filing Decision

A reader writes in, asking:

“Could you write something explaining the effects of taxes on the age you decide to begin Social Security. Especially since RMD’s may be delayed to age 72 under new legislation. Also, my state doesn’t tax Soc. Sec. benefits. Thanks.”

In most cases I have looked at, tax planning has worked out to be a point in favor of delaying.

The mechanism at work is that Social Security is, at most, 85% taxable. In contrast, distributions from tax-deferred accounts are usually fully taxable. And spending down your tax-deferred accounts in order to delay Social Security has the effect of increasing the portion of your lifetime income that is made up of (not-fully-taxable) Social Security and decreasing the portion of your lifetime income that is made up of (usually-fully-taxable) distributions.

And a similar thing is usually going on at the state income tax level. Only 13 states tax Social Security benefits, whereas a majority of states treat distributions from tax-deferred accounts as taxable income.

But, to be clear, the effect of taxes on the Social Security decision is very case-by-case. While the above effect is pretty broadly applicable, there could be any number of other factors that could point in the other direction. Almost anything that appears on a person’s 1040 could end up being a relevant factor in the analysis.

Ideally, the way to do the analysis (e.g., when comparing two possible claiming strategies) is to:

1. Use tax prep software (or other similarly fully featured tax planning software) to estimate the total household tax bill year-by-year under each claiming strategy that you want to test. (For a married couple, you actually want to estimate 3 different tax costs for each year for each claiming strategy: a scenario in which both people are alive, one in which only Spouse A is alive, and one in which only Spouse B is alive.)
2. Do a typical net present value calculation for each strategy, including the differences in tax costs as cash flows. For example, if you are comparing two strategies, and Strategy 2 has higher taxes by $1,000 in a given year, include that as a $1,000 negative cash flow for Strategy 2 that year. (Again, for a married couple, you would be doing three calculations for each year for each strategy — both spouses alive, only A alive, only B alive — then weighting each one by its probability, using a mortality table of your choosing.)

With regard to step #1, I would caution against using a spreadsheet or other similar DIY tax calculation. It’s very easy to accidentally fail to include a given credit/deduction/exclusion that would affect the analysis — especially when we consider state income taxes as well.

And of course it’s important to remember that all of this is just a projection. There are many unknowable factors involved.

Tangential note: my spreadsheet for doing step #2 of this analysis is what originally served as the starting point for the Open Social Security calculator. And it’s part of why I was surprised to realize that most (all?) other Social Security claiming calculators use a fixed life expectancy assumption in the calculation (i.e., assuming with 100% certainty that a person dies on a given date). Doing so is fine for an unmarried person, but for a married couple it significantly underestimates the length of time for which only one spouse will be alive. That really messes with the value of survivor and spousal benefits, and it also really messes up the expected tax cost calculation (because taxes change significantly once one spouse dies).