The Oregon Center for Public Policy (OCPP) has once again announced that that a study conducted by the Institute on Taxation and Economic Policy shows that Oregon state and local taxes hit poor and middle class harder than rich. This is an annual phenomenon.
But the conclusions of the study are almost certainly wrong, although the reality is nonetheless dismaying.
The weaknesses of the Institute on Taxation and Economic Policy Micro-simulation model are well known (it was roundly criticized by the discussants when it was presented at the National Tax Association meeting and the paper upon which the presentation was based was not selected for inclusion in the National Tax Journal). Its weaknesses derive from the arbitrary calculation of property-tax incidence and top coding of expenditures combined with the tiny sample sizes of the top two income categories in the sample (this may not be a significant weakness for the country as a whole, but it is for Oregon, which represents 1 percent of the total population in the sample).
The Institute on Taxation and Economic Policy estimates the property tax burden on renters based upon household rental expenditures. However we know this is wrong on the face of it. In the first place it leads to an estimate of property taxes paid on rental properties in Oregon that is greater than the total property tax levied on those properties. Moreover, it assumes that property taxes on rental properties are shifted forward to renters rather than paid by the property owners, which is inconsistent with what we know about the administration of property taxes on commercial properties in this state. A more realistic estimate of the property tax burden of the bottom quintile of the population is probably closer to 2 percent of income than 4.6 percent. Indeed, adjusting for this error would have the effect of reducing the total tax burden on the bottom quintile to about 6.6 percent, the second quintile to about 7.3 percent, the third to about 7.9 percent, and the fourth to about 8.8 percent.
The problem of top coding is harder to sort out, but we know the total property tax levy and total state income taxes paid as well as the adjusted gross incomes of the top quintile as well as the total income and property taxes they reported on their federal income taxes. The income tax results reported by the Institute on Taxation and Economic Policy are roughly consistent with these sources, but the property tax figures are not.
The conclusion I draw is that the overall property tax burden on the top quintile of Oregon households is understated considerably in the Institute on Taxation and Economic Policy study and the state income tax burden may be understated as well, although not by much. In which case, the average burden for the top quintile would be approximately 9 percent.
The Institute on Taxation and Economic Policy study has a couple of other weaknesses that deserves a look. It excludes user fees, which is now the main source of local revenue in Oregon, fines, and lottery revenues. These three categories of revenue now equal nearly 5 percent of total state disposable income. I would guess that the incidence of the first is probably progressive; we know that the incidence of the latter two revenue categories is regressive. Their effect deserves a visit. Finally, the underlying expenditure data used in the micro-simulation (not the income data) are also now over ten years old and should be updated.
Nevertheless, the Institute on Taxation and Economic Policy study’s results for excises are probably still reasonably accurate (at least for the first five income categories
reported) and quite informative. Folks should bear these figures in mind when they contemplate increases in consumption taxes. Unfortunately, this is where most state tax increases have been concentrated in recent years.
Moreover, I would stress that OCPP’s policy conclusions make a lot of sense, even if we adjust the Institute on Taxation and Economic Policy’s figures to make them more realistic. These include "making our tax system more fair by expanding the Earned Income Credit" and that "Oregon’s income tax brackets and rates are flat compared to the federal system." It is unconscionable that Oregon’s lowest tax bracket starts at $2,750 and its highest at $6,851 of earned income. The minimum should be at least $10,000 or $20,000 for a household filing jointly. That would leave us with a flat tax on incomes above those levels. Increasing that rate roughly .5 percent would offset the revenue loss of so doing. The effect would be to make the state income tax at least as progressive as the federal income tax and somewhat more progressive than total federal taxes.