How Regressive is Minnesota's Tax System? Very Regressive
Lower and middle income Minnesotans are paying a larger share of their income in taxes than wealthier earners. In fact, the state and local tax system in Minnesota is more regressive today than at any time in recent history, according to the 2009 Minnesota Tax Incidence Study (MTIS).
The 2009 MTIS focuses on tax information for 2006. The major objective of the MTIS is "to provide taxpayers and policymakers with important information on the equity or fairness of the overall distribution of Minnesota taxes." The MTIS is prepared every two years by the Minnesota Department of Revenue in response to a statutory requirement enacted in 1990.
The "Suits index" is a statistical tool used to measure the regressivity of a tax system. A tax system in which the effective tax rate (i.e., total tax as a percent of income) is similar across all income groups is said to be proportional. A proportional tax system has a Suits index of zero. A positive Suits index (i.e., an index with a value greater than zero) denotes tax progressivity (i.e., high income households pay a higher percentage of their income in taxes than do low income households). A negative Suits index denotes tax regressivity; the further below zero, the greater the degree of regressivity.
There is a trend toward greater regressivity over the history of the MTIS. As measured by the Suits index, Minnesota's tax system was more regressive in 2006 than at any time in the history of the MTIS. Furthermore, the increase in tax regressivity from 2004 to 2006 was greater than during any other two year period covered by the MTIS. From 2004 to 2006, the regressivity of Minnesota's tax system as measured by the Suits index approximately doubled from -0.024 to -0.053.*
Like many statistical concepts, the significance of the Suits index can be difficult to express. For example, precisely how regressive is a tax system with a Suits index of -0.053? How significant is a decline in the Suits index from -0.024 to -0.053?
There is no simple way of answering these questions because in a regressive tax system, the effective tax rate does not generally increase at a uniform rate from the lowest decile (i.e., the ten percent of households with the lowest income) to the highest decile (i.e., the ten percent of households with the highest income). However, it is possible to make some useful generalizations about the degree of regressivity associated with a Suits index of a particular value if we assume that the effective tax rate increases at a uniform or linear rate from lowest income to highest income households.
Work of this nature has been done by Dr. Paul Wilson of the Minnesota Department of Revenue. An example of a graph prepared by Dr. Wilson for presentation to the Minnesota Senate Tax Committee can be found here. (PDF) In this graph, Dr. Wilson assumes an average effective tax rate of 11.2 percent, which is the statewide average for Minnesota in 2006 based on the 2009 MTIS. Dr. Wilson then shows how effective tax rates vary at different incomes level for Suits indices of varying values.
Using a similar approach, it is possible to make generalizations about the significance of a -0.024 Suits index, the value of the Suits index in Minnesota in 2004, versus a -0.053 Suits index, the value in 2006.
The blue line in the graph represents a proportional tax system with a Suits value of zero. The total tax as a percent of income is identical at all income levels. Households in the lowest income decile are paying 11.2 cents of each dollar of income in state and local taxes-the same as households in the highest income decile.
With a Suits index with a value of -0.024 (the green line), households in the lowest income decile are paying 12.9 cents of each dollar of income in taxes, while households in the highest income decile are paying 10.8 cents. In other words, the lowest income households are paying 19 percent more in taxes per each dollar of income than are the wealthiest households. This represents significant tax regressivity.
With a Suits index of -0.053 (the red line), the disparity in taxes between low and high income households becomes even more stark. Households in the lowest income decile are paying 14.9 cents of each dollar of income in taxes, while households in the highest income decile are paying 10.3 cents. The lowest income households are paying 44 percent more in taxes per each dollar of income than are the wealthiest households.
In reality, tax burdens do not change at a uniform rate across income levels. (For example, due to data anomalies state and local taxes per dollar of income are over 2.3 times greater in the lowest income decile than in the highest income decile based on 2006 data.) However, assuming a uniform rate of change in effective tax rates across income levels allows us to make useful generalizations about the significance of the Suits index value and changes in that value over time.
Based on these generalizations, it is clear that Minnesota's tax system was significantly regressive in 2004 and became much more regressive by 2006. Projections in the 2009 MTIS show no significant reduction in regressivity from 2006 to 2011 based on current law.
Year-to-year changes in the regressivity of the state's tax system are largely driven by economic forces beyond the control of state government, although tax policy decisions clearly make a difference. Policymakers should do what they can to mitigate the regressivity of Minnesota's tax system by reducing reliance on regressive taxes, such as the property tax, and increasing reliance on progressive taxes, such as the income tax.
State policymakers-including Governor Pawlenty-have acknowledged that the state needs new revenue (in addition to spending cuts) to help balance its huge $6.4 billion structural deficit. This new revenue should come from progressive taxes so as to shift more of the state tax burden away from the state's poorest households to those with the greatest ability to pay.
*Based on "population decile" Suits indices as reported in the 2009 MTIS.
+The ETRs shown in both graphs were calculated using actual 2006 population decile income information with tax levels adjusted so as to produce (1) an average ETR of 11.2 percent and (2) an ETR that changes at a uniform rate from one population decile to the next.