The use of an incorrect tax base, such as the Michigan Tax Base, introduces systematic bias into the results. Because the Michigan Tax Base includes both the capital expenditure and its depreciation in the tax base, it biases upward the tax bases of industries that made large capital expenditures in the sample periods. Dividing tax liabilities by the (inflated) Michigan Tax Bases produces, then, tax burden ratios that are biased downward.
Thus, the indicated "effective tax rates" in the Treasury report underestimate the true tax burden. [31]
Because different industries rely on capital expenditures to a varying degree, and because different industries make large capital expenditures at varying times in the business cycle, this bias could produce very large swings away from the underlying true tax burden ratios. Therefore, the magnitude of the bias would change from year to year, meaning again that a reliance on a very short sample period would be clearly misleading.
In addition to the general downward bias on the tax burden ratios in the Treasury study, industries that made substantial capital investments relative to their Michigan Tax Bases during the sample period of the Treasury study seemed to have lower relative tax burden ratios than a proper comparison would have produced, while those who made little capital investment seemed to have higher relative tax burden ratios.
Biased Results from the Use of an Incorrect Tax Base
The 1985 Treasury Report relied on an incorrect measure of the tax base, which ignored capital expenditures. Thus, their analysis systematically biased downward their calculated SBT burdens. The actual tax burdens of Michigan industries were higher than those stated in the Treasury study, particularly for those industries making large capital expenditures in the sample period.