While I haven't received any of my tax forms for 2017, I can compute the values for each of the boxes based on past experience and available statements and pay stubs. My tax situation is fairly simple: 1 W-2, 3 1099-INTs, 2 1099-DIVs, and 1 1099-B (the gain was only ~$5). I itemize, only accounting for ~$1000 in charitable donations, which I value myself; property taxes; and mortgage interest. I file married jointly, and I have one young child.

Thus far, I have compared TurboTax, H&R Block, TaxAct, and Credit Karma. Concerning only federal for now, the last three predict the same refund, just above ~$1000, whereas TurboTax predicts a greater refund by $28.

I can't seem to find anything to suggest any input discrepancies between TurboTax and the others. Additionally, from what I can tell, the inputs are pretty straightforward -- no room for different interpretations. Last year, I compared TurboTax and H&R Block, and both returned the same values.

Therefore, my question is whether anyone knows of any subtle difference(s) in how TurboTax computes the refund, relative to other products. If not, I may go through and remove/add different components in an attempt to isolate the culprit.

  • 4
    My guess would be that one of the products may assume a credit or deduction based on history (did you load prior year returns into some but not all products), or other assumption. Can you see whether any other differences on the output pages exist? ie: Does your 1040 printout look exactly the same otherwise? Jan 8, 2018 at 2:43
  • 1
    I've narrowed it down to something that arises once the algorithm switches from standard deduction to itemized.
    – BaronFiner
    Jan 8, 2018 at 3:20
  • AGIs are identical?
    – BobE
    Jan 8, 2018 at 4:31
  • Yes -- culprit was 2016 state tax paid in 2017.
    – BaronFiner
    Jan 8, 2018 at 4:39

2 Answers 2


@grade-eh-bacon was right. It turns out the culprit was $118 state tax due for 2016 in 2017. Each year, I typically get a federal refund, but owe a bit of state tax, typically, around ~$20. In 2016, I owed more than usual.

For 2016, I entered the information into TurboTax, despite not using it. The free version of H&R Block I used then did not store the information for 2017 like TurboTax did. This is only an issue when one itemizes.


Print out the forms and go through them line by line. I'd expect to find one line somewhere where the $28 shows up on one and not the other, or where the two differ by $28. See what that line is. Either one of them is including something that the other isn't, or there's a difference in a calculation. Maybe TurboTax thinks you qualify for some deduction or credit that the others don't. Or one or the other is using an out-dated formula somewhere. I can't imagine how anyone could tell you where the discrepancy is coming from without looking at the detailed forms. You have to find the line containing the discrepancy and track it down from there.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .