Dept. of Rev. v. Butte Creek Associates I

I. INTRODUCTION This matter is before the court for decision after trial, with a stipulation as to certain facts. II. FACTS To set this case in proper context, it is useful to first lay out the relevant background. As evidenced by a series of decisions from this court and the Supreme Court, Plaintiff (the department) and various property owners have engaged in a long-running struggle to determine how best to assess low-income housing projects for property taxation. See Bayridge Assoc. Ltd.Partnership v. Dept. of Rev., 13 OTR 24 (1994), aff'd,321 Or 21, 892 P2d 1002 (1995); Piedmont Plaza Investors v. Dept. ofRev., 14 OTR 440 (1998) (Piedmont Plaza I), rev'd,331 Or 585, 18 P3d 1092 (2001) (Piedmont Plaza II); WilsonvilleHeights Assoc., Ltd. v. Dept. of Rev., 17 OTR 139 (2003),aff'd, 339 Or 462, 122 P3d 499 (2005). Those cases were litigated against a statutory backdrop in which real market value (RMV) was determinative of assessed value (AV). *Page 3 In 2001, the legislature passed House Bill (HB) 2204, which established a new taxation regime for low-income housing projects, which would apply at the option of the property owner. Or Laws 2001, ch 605. HB 2204 was codified at ORS 308.701 to308.724.1 The new regime offers owners of low-income housing projects (defined as "multiunit rental housing that is subject to a government restriction on use") the choice of either having their property assessed "under the ordinary methods of assessing property in this state" or having it assessed under ORS 308.707. ORS 308.704. ORS 308.707(4) provides that low-income housing projects shall be assessed at the lowest of the property's real market value, specially assessed value (SAV), or maximum assessed value (MSAV). MSAV is a function of the SAV. ORS 308.707(3). The SAV of a property is to be determined according to one of three methods described in ORS 308.712, chosen by the property owner. In all cases, the RMV of a property, determined under the general principles of ORS 308.205, if lower than the SAV or MSAV of the property, will be the basis for assessment. ORS 308.707(4). Defendant (taxpayer) operates a low-income housing project built in 1990, located in Jackson County (the county), and qualifying for the benefits of section 42 of the Internal Revenue Code and section 515 of the Federal Housing Act of 1949,42 USC § 1485 (1990). For the 2002-03 tax year, the first year in which taxpayer's property could be specially assessed according to the provisions of ORS 308.701 to 308.724, taxpayer chose to have its property specially assessed. Under ORS 308.712, taxpayer was therefore required to elect one of three methods "by which the specially assessed value of the property is to be determined." Taxpayer chose the method set out in ORS 308.712(1)(a) (the income approach method), which determines the property's SAV: "Through an annual net operating income approach to value that uses actual income and stabilized operating expenses that are based on the actual history of the property (if available) and a capitalization rate. The income, expenses and capitalization rate used must be consistent *Page 4 with the Uniform Standards of Professional Appraisal Practice [(USPAP)]and may be further defined by rules adopted by the Department of Revenue. Factors to be considered in setting a capitalization rate include the risks associated with multiunit rental housing subject to a government restriction on use, including but not limited to diminished ownership control, income generating potential and liquidity. The capitalization rate that is set pursuant to this paragraph must be equal to or greater than the capitalization rate used for valuing multiunit rental housing that is not subject to a government restriction on use." 1. Under ORS 308.724, the department must "prescribe rules implementing the provisions of ORS 308.712(1)(a)," which then become part of the formula used to determine SAV under ORS308.712(1)(a). Accordingly, any valuation under ORS 308.712(1)(a) must conform to both the provisions of that statute and the rules adopted by the department pursuant to ORS 308.724, at least to the degree that those rules do not conflict with the provisions of ORS 308.712(1)(a).2 The department has adopted rules to implement ORS 308.712(1)(a), which are found in OAR 150-308.712(3).3 Those rules state: "(3) The income approach method — For the initial year of special assessment, the assessor utilizes the property's actual income statements for at least the prior three years. Pro forma statements may be used for recently constructed properties. A combination of actual and pro forma statements may be used. "(a) The goal of the income approach is to determine the value of only the real property. No personal property value should be included. The assessor may remove personal property value by one of the following methods: "(A) Include revenues and expenses for both the real and personal property. After the net operating income has been capitalized, deduct the value of the personal property; or *Page 5 "(B) Remove all income and expense generated by the personal property assets prior to capitalization. "(b) In determining the SAV, no income should be included for government income tax credits or mortgage interest subsidies. "(c) The assessor must use actual income (revenues) and stabilized expenses. "(d) Actual revenues included are those which result from the operation of the property. They include the rent paid by tenants and any monthly rent subsidies. Also, rent for parking or other amenities must be included. Revenue not directly related to the property, such as interest income, should be excluded. "(e) Stabilized expenses are those that would be expected to be typical for the property; not those that reflect unusual or extraordinary circumstances. The assessor may use averages for the three years and may express expenses on a per-unit basis or as a percentage of revenue. Expenses for a particular year should be adjusted if they are atypical. The goal is to find the typical level of expenses. "(f) Expenses to include are those directly related to the operation of the property including, but not limited to, repairs and maintenance, utilities, government required tenant services, management and insurance. Certain expenses such as depreciation, mortgage interest, payments to developers and property taxes must be excluded. Reserves for replacements should be included, but any expense in the repair and maintenance category should be disallowed if it comes from the reserve account. "(g) The net operating income is determined from the above steps by subtracting the stabilized annual expenses from the actual annual revenues. "(h) The capitalization rate is estimated as follows: "(A) Factors to be considered in selecting a rate include the risks associated with multiunit rental housing subject to government restriction. These include diminished ownership control, income generating potential and liquidity. The assessor must also consider any other factors or risks typically taken into account when estimating a capitalization rate. *Page 6 "(B) The selected capitalization rate must be equal to or greater than the rate used by the assessor for similar unrestricted properties. "(C) To the selected rate, add the effective property tax rate for the code area where the property is located. This is the overall rate to use for capitalization. "(D) The value determined from the income approach is calculated by dividing the overall capitalization rate into the net operating income. This is the SAV. Notwithstanding the result of the calculation, the SAV of the real property land and improvements may not be less than $1,000 per dwelling unit." Using the income approach method set out in ORS 308.712(1)(a) and OAR 150-308.712(3), taxpayer determined that the SAV of its property was approximately $175,000. The department reached a different result, a SAV of approximately $508,850. Taxpayer appealed to the Magistrate Division, which found the property's SAV to be $270,073. The department appealed, arguing again that the correct SAV of the property is no less than $508,850. In turn, taxpayer argues that the property's SAV is no greater than $270,073. The sole issue on appeal is the SAV of taxpayer's property; neither party raises the issues of RMV or MSAV.4 The parties have stipulated to certain facts, which are recited where relevant in the court's opinion. III. ISSUE Under ORS 308.712(1)(a), what is the correct SAV for the property for tax year 2002-03? IV. ANALYSIS 2. A determination of SAV under ORS 308.712(1)(a) and OAR 150-308.712(3) requires calculation of taxpayer's *Page 7 actual income and stabilized operating expenses, and a capitalization rate. The stabilized operating expenses are subtracted from the actual income, resulting in taxpayer's net operating income (NOI). OAR 150-308.712(3)(g). Here, the parties have stipulated that taxpayer's NOI for the tax year at issue was $39,568. The NOI is then divided by the sum of the capitalization rate and the effective tax rate. OAR 150-308.712(3)(h). From that figure is subtracted the value of taxpayer's personal property, yielding the ultimate SAV of taxpayer's property. OAR 150-308.712(3)(a)(A).5 The parties have stipulated that the value of taxpayer's personal property was $10,000 for the 2003-03 tax year. Accordingly, the only issues remaining are the correct capitalization rate and effective tax rate. The court analyzes each rate in turn. A. Capitalization Rate In this case, both parties rely on appraisals that adopted a similar framework for determining the capitalization rate. First, the appraisers calculated an average capitalization rate for unrestricted housing that is similar to taxpayer's restricted housing in all respects other than the government restrictions (the base capitalization rate). That calculation was derived from comparable sales information. See, e.g., Appraisal Institute,The Appraisal of Real Estate 531-33 (12th ed 2001) (describing the generally accepted method for derivation of a capitalization rate based on comparable sales); Hope Village, Inc. v. Dept. ofRev., 17 OTR 370, 379, 381 (2004) (describing when resort to comparable sales of a related kind of property should be made). Second, the appraisers calculated an adjustment for the risk associated with restricted housing that is not associated with unrestricted housing. That adjustment was designed to account for the differences between the two types of housing. The general approach utilized by both appraisers, although not required by either ORS 308.712(1)(a) or OAR 150-308.712(3), is sound and is not inconsistent with the statute or the rule. It recognizes the difficulty in making direct *Page 8 comparisons between taxpayer's restricted housing and other restricted housing given the lack of free market transactions involving such properties. It then seeks to make a determination of value based on the information that is available and reliable, as required by USPAP. In reviewing the appraisals in this case, it is best to analyze each step in the appraisers' common framework in turn. 1. Base Capitalization Rate The department's appraiser, David Arrasmith, a deputy assessor with the county, utilized both a band-of-investment method and a comparable sales approach to calculate a base capitalization rate. Under the band-of-investment approach, Arrasmith found that the average capitalization rate for unrestricted housing otherwise similar to taxpayer's restricted housing was 9.06%. Under the comparable sales approach, he found an average rate of 8.35% based on an examination of four comparable sales. Giving equal weight to the 8.35% and 9.06% figures, Arrasmith ultimately concluded that the appropriate base capitalization rate was 8.7%. In contrast, taxpayer's appraiser, Mark Skelte, determined that a base capitalization rate of 9.25% was appropriate. Skelte based his appraisal on an analysis of six comparable sales, three of which were also utilized by Arrasmith. Nonetheless, the two appraisers found different base capitalization rates for each of those properties. Skelte attributed to one comparable, Alder Street, a capitalization rate of 8.98%, while Arrasmith rated it at 8.57%. Another comparable, Woodcreek, was rated at 9.31% by Skelte and at 9.01% by Arrasmith. A third comparable, Brookside, was rated at 9.26% by Skelte and at 7.59% by Arrasmith. Not surprisingly, each party attacks the analysis made by other party's appraiser and defends its own appraiser's analysis. On the one hand, the department attacks Skelte's appraisal for failing to take into account present indicators of income, and defends Arrasmith's appraisal as based on discussions with the property owners and research into average regional and national capitalization rates. On the other hand, taxpayer attacks Arrasmith's appraisal for failing to take into account expectations of future income, and defends Skelte's appraisal *Page 9 as based on actual data used by those who bought the comparable properties. The court finds neither appraisal wholly convincing. Skelte at one point testified that his own appraisal was "sloppy." For instance, although Skelte placed primary reliance in part on the Woodcreek property, he increased the NOI of that property by more than $25,000 over the NOI stated in the materials he relied on. That increase resulted in a capitalization rate for Woodcreek of 9.31%, instead of the 8.6% rate stated in the materials. Skelte's explanation at trial for the discrepancy was that he had made an inadvertent mistake and that the rate should have been 8.6% as stated in the materials. He explained that much of the comparable sales information included in his appraisal report was taken from past research and reports, and that he had not reviewed all of it for accuracy, reliability, or relevancy to taxpayer's property. In a related vein, Skelte's appraisal failed to include the materials he relied on to determine capitalization rates for four of his other comparable sales; consequently, the court cannot review whether Skelte made other mistakes similar to the one he made with Woodcreek. Arrasmith's appraisal, too, suffers from a lack of reliability. For instance, Arrasmith attributed to his comparable sales, which were unrestricted housing, some expenses that he derived from restricted housing, despite significant differences in the amounts of those expenses both between unrestricted housing and restricted housing, and between comparable restricted housing properties. That, and other questionable methods used by Arrasmith, render his appraisal less persuasive. Despite the problems with both appraisals, the court must make a finding as to which base capitalization rate is appropriate for use in this case. The court finds that an appropriate base capitalization rate in this case is 9.0%. That figure comports with Arrasmith's band-of-investment analysis, as well as the figures both appraisers derived from the comparable sales approach. Those comparable sales that were most like taxpayer's property, such as Alder Street, and those comparable sales most relied on by the appraisers, including Woodcreek, were or were meant to be rated by the *Page 10 appraisers at a rate lower than Skelte's ultimate rate of 9.25%, yet higher than Arrasmith's ultimate rate of 8.7%. In short, the evidence from all comparable sales, as well as Arrasmith's band-of-investment analysis, supports a base capitalization rate that lies between the rates stated in the two appraisals; that rate is 9.0%. 2. Risk Adjustment 3. Both parties agree that the base capitalization rate must be adjusted for the differences between restricted and unrestricted housing. See ORS 308.712(1)(a) (stating that the capitalization rate must reflect "the risks associated with [restricted housing], including but not limited to diminished ownership control, income generating potential and liquidity," and that the overall capitalization rate "must be equal to or greater than the capitalization rate used for valuing [unrestricted housing]"). However, the parties dispute the proper methodology for determining the size and direction of that adjustment. The department argues that the overall capitalization rate must take into account not only the restrictions associated with restricted housing, but also the benefits that government housing programs confer, including "budgeting procedures, nonrecourse loans, rental subsidies, reserve account requirements, and other features that result in a high degree of certainty that the NOI will be achieved."6 The question in this case is whether and how those benefits are to be considered in developing an SAV for taxpayer's property under ORS 308.712(1)(a). The department argues that the benefits often outweigh the restrictions, such that restricted housing generally involves less risk than unrestricted housing. Taxpayer argues that, to the extent that any benefits are to be considered, they are outweighed by the restrictions, such that restricted housing generally involves greater risk than unrestricted housing. The department urges the court to accept a novel theory of risk. First, the department looks to the Appraisal *Page 11 Institute's sound definitions of "risk." See Appraisal Institute, The Dictionary of Real Estate Appraisal 312 (3d ed 1993) (defining risk as the "probability that foreseen events will not occur"); Appraisal Institute, The Appraisal of RealEstate at 492 ("The anticipation of receiving future benefits creates value, but the possibility of not receiving or losing future benefits reduces value and creates risk."). The department then seeks to argue that any foreseen risk is no risk at all. As relates to this case, the department appears to argue that, because taxpayer knew that restricted housing is associated with diminished control, income potential, and liquidity, those factors create no risk for taxpayer.7 That view distorts the meaning of risk both in general and as applied to ORS308.712(1)(a). 4. The differences between restricted and unrestricted housing, including both benefits and restrictions, are taken into account by investors when they choose to purchase either one or the other kind of property. Those differences affect an investor's ability to respond to market changes; that ability is an element to be reflected in the concept of risk. See Appraisal Institute, TheAppraisal of Real Estate at 94-95 (describing various kinds of risk, including market risk and liquidity risk). In theory, if all conditions existing at the time of investment were to stay constant, then the department's approach might have value, at least as to any one property. But market conditions and investor needs do change, at which point factors such as control, income potential, and liquidity come into play. An investor who anticipates a need for quick cash, for example, will prefer a property that has greater liquidity. An owner of restricted housing may not be able to take advantage of increased rates in the rental market, but may be somewhat protected if general market rents decrease. Investors take those factors into account when analyzing the differences between restricted and unrestricted housing, and, ultimately, placing a value on the different levels of risk associated with the two kinds of properties. The *Page 12 department's methodology appears to focus on the risk to one property. The language of the statute makes clear that the legislature intended there to be a comparison of two property types: restricted and unrestricted. Thus, even though the benefits associated with restricted housing may be taken into account in determining a capitalization rate under ORS 308.712(1)(a), and although it is possible that a particular restricted housing property might have a lower capitalization rate than similar unrestricted housing properties, neither of those facts supports the department's theory of risk or its failure to analyze the relative or comparative risk between restricted and unrestricted housing. Instead, the requirement of ORS 308.712(1)(a) that any calculation of SAV must be guided by generally accepted appraisal principles, as reflected in USPAP, accurately reflects the true nature of risk, as described by the court. 5, 6. Apart from the theoretical definition of risk, Arrasmith's method of determining a risk adjustment for taxpayer's property is also completely unreliable. Arrasmith first concluded that restricted housing comprises a sub-market of all housing properties. He then compared taxpayer's property to similar restricted housing properties, in terms of NOI and financing terms, using comparable sales. That comparison resulted in negative risk adjustments of 1.39% (based on NOI) and 2.19% (based on financing terms), to be subtracted from Arrasmith's base capitalization rate.8 That approach was erroneous. The comparable sales relied on by Arrasmith were preservation transfers of the type described in Piedmont Plaza I, in which this court held that a "preservation transfer is clearly not open market value and is not a comparable sale." 14 OTR at 450. InPiedmont Plaza II, the department conceded to the Supreme Court that this court's holding in that regard was correct.331 Or at 591. This *Page 13 court finds it exceedingly odd, then, that the department would now seek to repudiate the position it took before the Supreme Court, reject this court's prior holding, and rely on preservation transfers as evidence of value. The court takes this opportunity to reiterate its finding in Piedmont Plaza I and therefore agrees with taxpayer that the sales relied on by Arrasmith are not appropriate to consider in determining a risk adjustment for the base capitalization rate under ORS308.712(1)(a).9 Accordingly, the court gives no weight to Arrasmith's determination of a risk adjustment.10 That conclusion is of special significance given that the department bears the burden of proof in this case. ORS 305.427. Skelte, on the other hand, compared ownership of restricted housing to the holding of a partial interest in property or a partnership owning property, noting that the latter situations also involve diminished control and marketability, and often diminished income potential as well. That approach has limitations due to the differences in the property interests involved, but is reasonable because it accounts for many of the risks associated with restricted housing, including two of the three risks specified by ORS 308.712(1)(a): those of diminished control and marketability or liquidity. Further, the approach attempts to use market evidence of at least some sort. Using that approach, Skelte included nine comparable sales in his report, concluding from that data that the risk adjustment should be a positive 35%, *Page 14 that is, that the base capitalization rate should be multiplied by 1.35.11 The court would adopt Skelte's risk adjustment as fact, but two factors cast doubt upon aspects of his appraisal. First, as the department points out, there is no evidence to support the information contained in the risk adjustment portion of the appraisal report. The only information before the court is Skelte's report and testimony; there are no records of the comparable sales. Without such evidence, it is difficult to review meaningfully whether the appraisal report contains complete, accurate, and reliable information, although the court notes that Skelte was subject to cross-examination on his factual bases. Second, even if the court takes as true all facts contained in the report, it appears that Skelte drew a questionable conclusion: his risk adjustment of 35% is too high. Skelte concluded in his report that the high end of the risk adjustment scale was between a positive 31% and 84%, and that the low end of the scale was at a positive 5.9%. The property Skelte found most comparable to taxpayer's property indicated a risk adjustment of 26.3%.12 Two other properties that Skelte considered strong indicators of an appropriate adjustment reflected an adjustment of 5.9% and 20%-40%. Other properties on which Skelte also relied indicated a risk adjustment between 20% and 30%. Given the information contained in Skelte's appraisal report, the court finds that Skelte's risk adjustment of 35% is not well supported by the evidence. Instead, the court finds that the appraisal supports a risk adjustment of 26.3% and adopts that figure for this case. 7. To derive the overall capitalization rate, the base capitalization rate must be increased or decreased by the risk adjustment. Here, the 9.0% base capitalization rate must be *Page 15 increased by a factor of 26.3%, that is, it must be multiplied by 1.263. The result is a capitalization rate of 11.367%. That result is greater than Arrasmith's base capitalization rate of 8.7% and Skelte's base capitalization rate of 9.25%, and is therefore consistent with the command of ORS 308.712(1)(a) that the capitalization rate "be equal to or greater than the capitalization rate used for valuing [unrestricted housing]." B. Effective Tax Rate ORS 308.712(1)(a) does not mention an effective tax rate, or any tax rate at all; it only requires that the capitalization rate take into account the various factors and risks described above, as well as meet the requirements of USPAP and the department's rules, which ORS 308.724 specifically requires the department to prescribe. The "effective tax rate" language is found in OAR 150-308.712(3)(h)(C), which states: "To the selected [capitalization] rate, add the effective property tax rate for the code area where the property is located. This is the overall rate to use for capitalization." The department has not specified by rule how the term "effective tax rate" is to be defined or determined. 8. The purpose of adding the effective tax rate to the capitalization rate is to account for the fact that property taxes are not included in the NOI stipulated to by the parties. Appraisal Institute, The Appraisal of Real Estate at 513. That is because the goal of discovering the SAV is to determine the amount of property taxes owed on the property. It is, of course, impossible to factor property taxes into the NOI without knowing the amount of those taxes. Nonetheless, property taxes must be included in the valuation somewhere because they reduce the property's income. With that background in mind, it must be remembered that properties in Oregon are not generally taxed on their RMV, or even on their SAV, but on their AV. See ORS308.146 (so stating for most properties); ORS 308.707(4) (so stating for restricted housing); Hope Village, 17 OTR at 376-77 (describing the post-1997 regime of property taxation in Oregon). Accordingly, as both parties admit, some adjustment must be made to the millage rate, if it is calculated in terms of RMV or SAV, to account for the difference between RMV and SAV, on the one hand, and AV on the other hand. See Allen v. Dept. of *Page 16 Rev., 17 OTR 248, 259 n 12 (2003) (noting that "adding the millage rate to the cap rate" "does not best approximate the proper tax burden because of the impact of [Measure 50]" and that "the nominal tax rate may not be the actual tax rate"). 9. Here, the parties differ on the appropriate method to use in making the necessary adjustment; their arguments amount to a dispute over the proper interpretation of the term "effective tax rate" as used in OAR 150-308.712(3)(h)(C). In construing administrative rules, the court must uphold an agency's interpretation of its own rule so long as that interpretation is "plausible" and "cannot be shown either to be inconsistent with the wording of the rule itself, or with the rule's context, or with any other source of law." Don't Waste Oregon Com. v. EnergyFacility Siting, 320 Or 132, 142, 881 P2d 119 (1994). In this case, then, the court must uphold the department's interpretation of OAR 150-308.712(3)(h)(C) so long as that interpretation is plausible and not inconsistent with applicable law. Relevant to that inquiry are two stipulations made by the parties. "The 2002 tax rate applicable to assessed value or specially assessed value in the code area of [the property] is 1.7 mills, such rate calculated by dividing the tax collector's assessed value into the amount of taxes billed for tax year 2002-03." The changed property ratio (CPR) "for 2002 applicable to properties in the same class as [the property] is .82." Taxpayer contends that the stipulated millage rate, a tax rate of 1.7%,13 is the effective tax rate, because it includes any necessary adjustment for Measure 50. The department, on the other hand, contends that the millage rate must be multiplied by the CPR, here a stipulated .82, to adjust for Measure 50. That calculation results in an effective tax rate of 1.39%. The court finds that the department's approach is not plausible and, therefore, cannot sustain it. There are at least two plausible ways of making the adjustment necessitated by Measure 50. One method would be to calculate a tax rate by dividing the tax collector's RMV into the amount of *Page 17 taxes billed for the tax year. See International Association of Assessing Officers, Property Assessment Valuation 241 (1977) (describing that method). That tax rate would then need to be adjusted in order to achieve the effective tax rate, for instance by multiplying it by the CPR. A second method would be to calculate the effective tax rate directly by dividing the tax collector's AV, and not the RMV, into the amount of taxes billed for the tax year. That approach sidesteps the need for an additional adjustment because it already accounts for the Measure 50-created difference between RMV and AV, the same difference reflected in the CPR.14 Here, the parties stipulated to a millage rate of 1.7% based on AV. Taxpayer's adoption of that rate as the effective tax rate, therefore, follows the second approach outlined by the court. The department, on the other hand, seeks to multiply the 1.7% rate by the CPR to achieve the effective tax rate; that method is implausible because it conflates the two approaches outlined by the court. As already pointed out, the parties' stipulated rate of 1.7%, based on AV, already accounts for the same difference between AV and RMV that is reflected in the CPR; no adjustment is necessary. To then *Page 18 multiply that rate by the CPR would be to account for the CPR twice, grossly overstating the actual difference between the RMV and AV of properties similar to taxpayer's in taxpayer's code area. Had the parties' stipulated rate been based on RMV, the department's CPR adjustment would be proper. That was not the case, however. Because the department's interpretation of OAR 150-308.712(3)(h)(C) is implausible, it cannot be sustained. The court determines that the proper construction of the department's rule as applied to this case results in an effective tax rate of 1.7%. C. Specially Assessed Value Once the capitalization rate and effective tax rate are known, the next step in determining a property's SAV is to add those two rates together to arrive at an ultimate denominator for the SAV equation. OAR 150-308.712(3)(h)(C). Arrasmith added his capitalization rate of 8.7% to his effective tax rate of 1.39% to determine a denominator of 10.09%. Skelte added his capitalization rate of 12.49% to his effective tax rate of 1.7% to determine a denominator of 14.18%.15 The court, however, has found that a capitalization rate of 11.367% is appropriate and that the effective tax rate is 1.7%; accordingly, the correct denominator is 13.067%. The next step is to divide the stipulated NOI, here $39,568, by 13.067%, OAR 150-308.712(3)(h)(D), which yields an unadjusted SAV of $302,808. However, the value of personal property must be subtracted from that figure to achieve the correct SAV of taxpayer's property. OAR 150-308.712(3)(a)(A). In this case, the parties have stipulated that taxpayer's personal property is worth $10,000. Accordingly, the SAV of taxpayer's property is $292,808. V. CONCLUSION The court finds that the correct base capitalization rate is 11.367%, the effective tax rate is 1.7%, and the SAV of taxpayer's property under ORS 308.712(1)(a) is $292,808. Now, therefore, *Page 19 IT IS DECIDED that the specially assessed value of Defendant's property under ORS 308.712(1)(a) is $292,808. 1 All references to the Oregon Revised Statutes (ORS) are to the 2001 edition. 2 Taxpayer does not argue that any of the department's rules adopted under ORS 308.724 are in conflict with the provisions of ORS 308.712(1)(a). 3 All references to the Oregon Administrative Rules (OAR) are to the current edition. See Wilsonville Heights Assoc., Ltd. v.Dept. of Rev., 339 Or 462, 470 n 8, 122 P3d 499 (2005) (referring to the current edition of the OAR based on OAR 150-305.100-(B)). 4 It is, of course, possible that the RMV of taxpayer's property will be lower than its SAV. RMV would be determined in accordance with the methodology set out in cases such asWilsonville Heights Assoc., Ltd. v. Dept. of Rev., 17 OTR 139 (2003), aff'd, 339 Or 462, 122 P3d 499 (2005). Of course, the MSAV of taxpayer's property will be lower than the SAV for the tax year at issue. See ORS 308.707(3)(a) (describing method for determining the MSAV of a low-income housing project in its first year of special assessment). In this case, as RMV has not been put at issue, the AV of taxpayer's property will be the lower of the property's MSAV determined under ORS 308.707(3)(a) or its RMV as stated on the rolls. ORS 308.707(4). 5 Another method is to subtract the income and expenses associated with a taxpayer's personal property before calculating NOI. OAR 150-308.712(3)(a)(B). The parties in this case, however, chose the method described in OAR 150-308.712(3)(a)(A). 6 Both the benefits and restrictions involved with restricted housing are discussed generally in Bayridge, Piedmont Plaza, and Wilsonville Heights. In this case, there was little, if any, evidentiary support as to how budgeting procedures, nonrecourse loans, and other features of restricted housing would, in fact, affect risk. 7 One of the major weaknesses of the department's approach is that it focuses almost exclusively on the risk of not achieving a stated level of NOI with a particular restricted housing property. That focus ignores the factors of control and liquidity, and discounts the possibility that a greater level of income could be achieved. 8 In other words, Arrasmith concluded that restricted housing similar to taxpayer's involves less risk than unrestricted housing similar in its physical characteristics to taxpayer's property. Because ORS 308.712(1)(a) requires that the capitalization rate for restricted housing be equal to or greater than that for unrestricted housing, Arrasmith did not make a risk adjustment. Instead, he concluded that the appropriate capitalization rate for taxpayer's property is 8.7%, the same as the base capitalization rate. 9 It is appropriate to consider actual market evidence, if such exists, in making adjustments to the base capitalization rate in order to establish the overall capitalization rate. That is a technique generally contemplated by USPAP, and the department agrees, in its written submissions, that market evidence is preferred. Arrasmith erred not by failing to use a market approach, but by failing to locate, and therefore use, market evidence. The evidence from preservation transfers is so compromised by nonmarket governmental policies, goals, and actions, that it cannot be considered market evidence. 10 The court notes that Arrasmith's utilization of comparable sales of restricted properties (the preservation transfers) to determine a risk adjustment contradicts the rest of his appraisal methodology. If there were reliable information about free market transfers of restricted housing similar to taxpayer's, then the appropriate capitalization rate for such housing could be determined directly and there would be no need to calculate either a base capitalization rate for unrestricted housing or a risk adjustment to equate unrestricted housing with restricted housing. Arrasmith could simply base his calculation on comparable sales of restricted housing and use the lower of that rate or the rate developed for the unrestricted housing market. 11 In his appraisal, Skelte relied on comparable sales of minority interests that occurred primarily between 1993 and 1996. That information is less reliable than more recent information, as taxpayer is well aware, given its criticism of Arrasmith's appraisal for the same fault. Nonetheless, that both parties' appraisers relied on older sales underscores the difficulty in finding complete, reliable, and accurate information in this context. In such situations, reasonable results based on sound methods and the best evidence available are persuasive. 12 Although Skelte's appraisal report indicates a risk adjustment of 23%, he testified at trial that the number should be 26.3%. 13 Although the parties stipulated to a millage rate of 1.7 mills, their filings, as well as the appraisal reports and the rest of the evidence, indicate that the actual millage rate intended by the parties is 17 mills, or a tax rate of 1.7%. 14 At one point in its post-trial brief, taxpayer, relying on the Appraisal Institute's The Dictionary of Real EstateAppraisal, recognizes that the effective tax rate is equal to the nominal tax rate multiplied by the assessment ratio, which is the fractional relationship that a property's AV bears to its RMV. Taxpayer is incorrect in stating that the assessment ratio only "refers to the situation where all property is assessed at a certain percentage of market value." Although that may be the case under different taxation regimes, in Oregon the assessment ratio can be approximated by using the CPR, at least under the broad authority granted the department under ORS 308.724. After all, the CPR is the same "ratio, not greater than 1.00, of the average maximum assessed value to the average real market value of property in the same area and property class as the specially assessed property" that is used to calculate the property's MSAV. ORS 308.707(3)(a). Cf. ORS 308.153(1)(b) (defining CPR in the context of properties not subject to special assessment as "the ratio, not greater than 1.00, of the average maximum assessed value over the average real market value for the assessment year"). Use of the CPR takes a large population approach to estimating the actual tax liability due on taxpayer's property. Alternatively, and consistently with the statute and rule, the effective tax rate could be measured on a more individualized basis, especially for unusual properties. Additionally, taxpayer correctly points out that sometimes the AV of a given property is the same as its RMV, in which case the effective tax rate for that property is the nominal tax rate. Nonetheless, the question is not what method might be best in any given case, but whether the department's interpretation of its own rule is plausible and not inconsistent with the statute, rule, and other applicable law. 15 Skelte alternated between using rounded and nonrounded numbers. *Page 20