Department of Revenue: 2023 tax-incidence study shows data limits likely overstate lowest-income burden

The Minnesota Senate Taxes Committee heard a presentation on April 8 from Eric Willette, director of tax research at the Minnesota Department of Revenue, on the department’s 2023 tax-incidence study, which analyzes who ultimately pays Minnesota state and local taxes.

Willette said the study covers taxes within the state code and excludes federal taxes and taxes paid to other states or countries. He reported that the individual income tax was the largest source of revenue in 2023 (about $15.3 billion) and that, overall, the department estimates roughly 50.6% of state and local taxes collected in Minnesota are exported outside the state. The department constructs a household database by combining income-tax filers, property-tax refund filers and administrative records from human-services programs to represent non-filing households; the sample contains about 400,000 records scaled to represent approximately 3 million tax-filing units.

A focal point of committee questioning and Willette’s presentation was the study’s finding that the lowest population decile shows a much higher effective tax rate than other deciles. Willette told the committee the 1st decile result should be treated with caution for two main reasons. First, a small number of households in that decile report very large negative business income (he cited roughly 9,000 households with average negative income around $112,000), which distorts the decile averages. Second, the department’s income measure omits in-kind benefits such as SNAP, housing assistance and medical assistance; Willette estimated roughly $1.5 billion in easily identifiable non-cash supports that—if included—would materially reduce the measured effective tax rate for the lowest decile.

Committee members pressed Willette on how the department represents non-filers and the reliability of administrative program data. Willette said the department augments filing records with human-services payment records and scales the resulting dataset to match population estimates, acknowledging limits in verification protocols for some benefits but saying the department relies on those administrative data sources for coverage of households that do not file income tax returns.

Willette also reviewed distributional measures including the Suits index. He said Minnesota’s overall Suits index is close to proportional (near -0.029 by the department’s measure) and that, aside from the lowest decile anomaly, the system is relatively flat across deciles. He identified the income tax, estate tax and property-tax refunds as the most progressive items and sales, motor-fuel and tobacco taxes among the most regressive. Willette noted that policy changes enacted in 2023—such as a child tax credit expansion and other refundable credits—made the income-tax system more progressive in isolation, but that broader economic changes since 2021 (particularly the fall in capital-gains receipts and the end of pandemic-era income supports) were responsible for many shifts in measured incidence.

Willette said the study includes a projection (production) year of 2028 to show how enacted changes and ongoing local taxes might affect incidence in the medium term, and he offered to provide additional breakdowns on homeowners versus renters by decile and on the department’s rough estimates of the value of non-cash benefits.

The committee did not take formal action on the report; members thanked Willette and said they may revisit the material at a future meeting.

The department’s full tax-incidence report and the slides shown to the committee are available from the Minnesota Department of Revenue for members seeking tables and underlying assumptions.