Nearly half of all tax returns filed are simple enough that they could be automatically, and accurately, pre-populated by the Internal Revenue Service, according to a recent study outlined in the
To come to this conclusion, the researchers examined a random sample of 344,400 individual income tax returns for 2019 constructed by the Statistics of Income Division of the IRS, a set which included information from nearly every line or box on Form 1040 and its schedules.
The researchers first checked the sampled returns for tax situations that would clearly result in an inaccurate pre-populated return, like reporting nontrivial amounts of income, deductions or credits that the IRS does not observe on the informational return.
They then simulated pre-populated 1040s line by line using only the taxpayer's 2019 information returns and their 2018 Form 1040 if they filed one. They then checked the pre-populated returns against what was actually reported to see if the calculated tax liabilities lined up, within a $100 tolerance threshold.
The results showed that between 41% and 48% of returns could be accurately pre-populated. Accuracy went down, however, as income went up. The researchers said these failures are largely driven by higher rates of itemized deductions, which cause a divergence between pre-populated and actual returns. This would suggest, said the study, that pre-population would be particularly successful for taxpayers who are single, young, and lack dependents. However, the study also pointed out that in 52 to 59% of cases where the pre-population was not accurate, the issue could be fixed with just one change or one additional schedule.
While asserting that pre-populating returns would save people significant amounts of time and money, the researchers said it would not be a casual thing to enact. Such a program would require the IRS to receive information returns earlier to avoid refund delays; the agency would also need to work fast to process and match information returns to generate the pre-populated ones. According to the researchers, that would mean that some of the reduced filing burden among individual taxpayers would be shifted onto information reporters and tax administrators. The study also mentioned the risk that people might simply accept whatever liability is listed as the default and not correct inaccurate information; further, since the pre-population would come from what is reported, such a program might incentivize people to not report things at all.
Nonetheless, the researchers said pre-populating returns would save a significant amount of time and money for taxpayers.
"We find that pre-populated returns have the potential to substantially reduce filing burdens for a large majority of taxpayers using existing information available to the IRS. While implementation details are challenging, we hope that this exploration will inform the discussion of innovative policies regarding pre-populated tax returns," said the study's conclusion.
