Oregon's elected officials should give us a chance to restore the property tax system, per-haps with a few improvements, that Measures 5 and 47 swept away. Granted, Oregon's old system of property taxes, local responsibility, and voter choice had problems, but they were fixable problems. The new system has problems too, really big problems. But the only way they can be fixed is by changing the system.

Of course, some well-informed observers don't seem to hold voter choice in very high esteem. The other day I went to a conference on the Initiative process in Oregon. One of the speakers, Bill Lunch, asserted that voters prefer lower taxes to higher taxes and more services to fewer services (which seems perfectly rational to me). He went on to say that voters (unlike their elected representatives) are inclined to ignore the discrepancy between these preferences and heedlessly vote to cut taxes and to increase spending. He pointed to Measures 5, 11, and 47 as evidence (47 cut taxes; 11 increased spending, and 5 promised to do both). Most of the audience nodded in agreement. Nevertheless, powerful evidence suggests that Mr. Lunch is wrong: between 1925 and 1995, Oregon voters imposed increased taxes on themselves in thousands of local tax levy elections.

Indeed, to my way of thinking, the difference between the behavior of voters in local levy elections and in statewide initiatives like Measure 5 is perfectly sensible. We, the people, behave responsibly only where we are, in fact, responsible, where we are directly affected by the consequences of our choices. Where local property taxes and user fees are concerned, the link between what we pay for and what we get in government services is direct and unambiguous. We can review and challenge our property tax burdens, compare them to those of our neighbors, and compare them directly to the benefits we receive. Furthermore, we can easily identify the property tax base; and we know that the tax is hard to evade. In contrast, no visible relationship exists between most of the other taxes we pay and the services federal and state governments provide, let alone the services we get from them.

Moreover, we know that the costs of property tax administration are much lower than the costs of administering sales and income taxes when our compliance costs are taken into account.

Nevertheless, many Oregonians voted for Measure 5, which reduced local property taxes. At the same time, they abandoned a voter controlled system that had worked fairly well for more than a hundred years. They replaced it with one that reduces governmental accountability and local autonomy and that greatly increases the authority of state officials and the dead hand of one-size-fits-all government. In Multnomah County, for example, property taxes now account for only 20 percent of the combined revenues of all local jurisdictions -- county, cities, schools, community colleges, other special districts, and regional authorities; intergovernmental aid accounts for a whopping 34 percent.

Obviously, I think that most objections to the property tax reflect a misunderstanding of how it actually works, otherwise why bother trying to explain it? Paradoxically, where the property tax is concerned, the people who have the worst understanding are the people who should be best informed -- opinion makers, lawyers, judges, and other public officials. Most property taxpayers I have talked to seem to understand how property taxes work. They know that decisions about property taxes and service levels directly influence the worth of their homes. Many understand that local action is justified if, and only if, government thereby increases the value of their property (which automatically means that the action also increases citizen welfare more than it increases tax obligations). Moreover, where local government actions increase citizen welfare and where assessments truly reflect the market values of the underlying assets being taxed, the beneficiaries of government action also pay for it. This sounds pretty good to me, but Bill Sizemore, and many otherwise very smart people just don't get it.

I suspect that understanding of the property tax breaks down for two reasons. First, assessments often do not reflect the economic value of the asset being taxed. Two-thirds of all the land in the state are excluded from the property tax base and much of the rest is under or arbitrarily assessed. Second, understanding how property taxes and service levels are capitalized into housing prices requires an appreciation of high school algebra. Many leaders of the bar, government, and the press chose their careers in part because they were a lot better at manipulating words than numbers.

So how does capitalization work? Imagine two towns, Alpha and Beta. They are identical except that Alpha has good parks and Beta does not. Some people value parks, and will choose to live in Alpha to be near them. Hence, the demand for and the price of land in Alpha will be greater than in Beta. The value of parks has thereby been capitalized into land values. More formally, "capitalization" is the process by which a stream of benefits or costs is converted into asset prices. Benefit streams increase asset values, and cost streams decrease them. If the benefits and costs flowing from an asset are constant over time, the following formula shows how capitalization works:

(1) P = B / i.

This formula says that the price of the land (P) is equal to the annual net benefit flow from the piece of property (B) divided by an appropriate discount rate (i).

Since the rent must cover both the tax and the normal return on the asset, a tax rate (t) on the land price is capitalized as:

(2) P = B / (i + t).

As t increases, P decreases, but B is unaffected. Nor can a local government that relies on the property tax shift the cost of capital improvements or current operations into the future by borrowing. Municipal debt is also capitalized, thus land prices are decreased by an amount equal to the present value of future property tax payments.

The capitalization of locally provided services and taxes into land values has been ex-tensively tested by economists and generally upheld. For example, William T. Bogart and Brian A. Cromwell, in an article entitled "How Much More is a Good School District Worth?" reported finding that where other things were held equal (including taxes and school spending per pupil), the reputation of its schools could add up to $50,000 to the average price of homes in a district. Another example: A. Quang Do and C. F. Sirmans, in an article entitled "Residential Property Tax Capitalization," said they had found that where other things were again held equal, each dollar paid in property taxes reduced housing prices about $25. The point is housing prices, local government services, and property tax rates are interdependent -- a point that should be obvious to homeowners who watched what happened to the market valuation of their homes following the enactment of Measure 5.

The interdependence of housing prices, services, and property tax rates has a direct bearing on one of the main objections to property taxes. Polls show voters believe that property taxes are unfair. It is not clear why. I have heard Norma Paulus, state superintendent of Public Instruction, describe as capricious the differences in municipal and school district spending and property tax rates. That is not true. Others complain that property taxes have no relationship to income: middle-class homeowners, struggling families and poor widows, and the rich all pay the same tax. That is true only if they all live in the same neighborhood and own similar properties and therefore receive similar benefits from their properties (measured in terms of what someone else would be willing to pay for those benefits, which is, by inspection, less than they would be willing to sell them for). This sounds to me like a benefit, not a cost, of the property tax.

The problems with the property tax as it was constituted prior to Measure 5 were real, but they reflected the way the tax was administered, not the tax itself. The first problem is that the property tax can be a real hardship for the elderly, especially those who are retired on fixed or reduced incomes. Most of oldsters own their own homes, many without mortgages. They must, therefore, pay large and inconvenient lump sums once or twice a year or be kicked out onto the streets. In contrast, those of us with mortgages find payment simple and convenient. Our bank pays the tax and, together with our mortgage payments, our taxes are automatically deducted from our monthly paychecks. Moreover, both the interest on our mortgage and our property tax payment are deductible from our income taxes. Government could easily find a way to make paying or deferring property taxes as cheap and convenient for the elderly, or indeed for anyone without a mortgage, as it is for the rest of us. Given the voting power of the elderly, it is amazing that our elected officials haven't already done so. However, in this instance, as in so many others, local responsiveness has been prohibited by state mandates.

The second problem with property taxes is that assessments really are unfair. Nearly two thirds of the state is not taxed. This includes all publicly owned land, plus all land occupied by religious, educational, and charitable institutions. Special exemptions are given to some property owners, including hightech companies, farmers and others. More importantly, commercial property is just not assessed the same way as private homes are. Assessments on private homes reflect the market value of the property and its improvements, including capitalization of the property tax itself; assessments on commercial property reflect whichever is the less of the capitalized value of the property's rental stream or its replacement value minus depreciation. Hence, when Measure 5 reduced tax rates, housing prices jumped and so did their assessments, but commercial assessments were largely unaffected, causing a substantial but probably unintended shift in the burden of the property tax from commercial property owners to homeowners.

Moreover, the way property is taxed in Oregon probably discourages efficient property development. Taxing improvements actually discourages people from maintaining or fixing up their property. Houses in poor condition, for example, were given a discount on their assessed value. Homeowners who maintained their homes were assessed at the full rate.

Again this issue could have been fixed by focusing the property tax on land -- away from improvements -- and subjecting all land to the property tax, regardless of ownership or current use. In that case, market prices could easily be used to assess all properties (and would deal fairly with difference in value that arise out of zoning, use, and development restrictions or convenants). Moreover, taxing land rather than improvements would mitigate the incentives to underutilize real property. For example, a March 1997 study by Wallace E. Oates and Robert M. Schwab showed that when the city of Pittsburgh, Pa., restructured its system of property taxation to one in which land was taxed at more than five times the rate on structures, it experienced a dramatic increase in building activity, far in excess of the other major cities in the region. This building boom occurred primarily in the city; it did not extend to the rest of the metropolitan area. In addition, it was a boom mainly in commercial construction; the residential sector experienced only a modest expansion, although maintenance levels were generally improved.

Instead, Measure 47 tried to fix this problem by freezing property tax assessments (as well as rates) except at transfer of title. Several states have this system, and evidently it creates problems of its own. In California, where the system has been in effect since 1978, it is not uncommon for one property to have an effective tax rate five times as high as its virtually identical neighbor. In Indiana, where the system has been in effect a lot longer but where property values have generally increased more slowly than in California, these discrepancies are less scandalous. Nevertheless, considerable support exists for returning the system of assessment to one based on market value. Unfortunately, the beneficiaries of Indiana's current system, the elderly, because they have stayed put, and owners of premium priced properties, because their real estate has experience the greatest increases in market valuation, now have a deeply vested interest in the status quo. If we don't change this system fairly soon, we are likely to be stuck with it for a long time.

The third problem with our old property tax system was that it permitted excessive variations in spending per pupil among school districts. This was primarily a problem of perceived unfairness. Differences in per pupil spending are only weakly related to student learning, if at all. In other states, equalization of per pupil spending and, in a few cases, inverse inequalities in school spending have not reduced inequalities in relative student performance, although the evidence suggests that where equalization has been associated with increased state financial responsibility, performance has fallen overall. That spending levels are generally unrelated to school quality is, perhaps, best illustrated by the fact that Measure 5 did not cause house prices to fall even in the highest spending school districts.

Nor were large differences in per pupil spending levels ever a truly serious problem in Oregon schools. Even when a third of the school districts in the state spent more or less than one standard deviation of the statewide average, those districts enrolled less than one percent of the students in the state, and many of the so-called low spending districts received uncounted timber tax or impacted area aid. We could have dealt (and at various times in the past did deal) with this problem quite satisfactorily by means of a well-designed foundation aid program. The past failures of foundation programs resulted from the state legislature's unwillingness to establish a high enough foundation amount and computational tax rate or to guarantee increases in those measures commensurate with increases in inflation. Of course, these facts do not exactly increase my confidence in the decision to place full responsibility for school spending levels in the hands of the state legislature and responsibility for the distribution of education funding in the hands of state bureaucrats -- the other set of consequences that derives from Measure 5.

A fair amount of evidence suggests that most Oregonians are not enamored with the consequences of Measures 5 and 47. That is what the polls show. Measure 5 passed by a razor thin margin; initiatives like it had been proposed and rejected over and over again by Oregon voters for years, before it finally passed. Maybe the voters were just trying to send a message about taxes in general. If that is the case, this legislature got the message -- they are not going to increase your taxes. But there is more. It would be possible to repeal Measures 5 and 47, fix the property tax system, restore local authority, provide fair funding to the schools without requiring absolute equality, and maybe cut some income taxes. Wouldn't that be better?





Fred Thompson

Grace and Elmer Goudy Professor

of Public Management and Policy

Atkinson Graduate School of Management

Willamette University, Salem, OR 97301


(503) 370-6228, FAX (503) 370-3011, Messages (503) 370-6440