Rialto Capital Advisors, LLC v. Marion County Assessor

IN THE OREGON TAX COURT MAGISTRATE DIVISION Property Tax

RIALTO CAPITAL ADVISORS, LLC ) and RSS JPMCC2012LC19-OR SCL, LLC, ) ) Plaintiffs, ) TC-MD 190092G ) v. ) ) MARION COUNTY ASSESSOR, ) ) Defendant. ) DECISION

This case concerns the 2018–19 real market value of the Salem Center mall (exclusive of its four anchors) and an adjacent movie theater. Plaintiffs’ trial counsel was Alex Robinson of CKR Law Group; Defendant’s was Scott Norris, Assistant County Counsel. Plaintiffs’ witnesses were Dana Vugteveen, the mall’s manager; Thomas Dobrowski, vice chairman in Newmark’s Capital Markets group; and Brian Booth, MAI, executive director at Cushman & Wakefield of Oregon. Defendant’s witness was Steven Tucker, Oregon Registered Appraiser. Plaintiffs’ Exhibits 1, 2, 3, and 5, and Defendant’s Exhibits A to G were received into evidence.

I. STATEMENT OF FACTS

A. Property Overview

On the assessment date, the portion of Salem Center under appeal included the concourses, mall shops, and major tenant space within two buildings on adjacent city blocks— Salem Center North and Salem Center South. (Ex 1 at 22.) Each building contained an anchor department store not owned by Plaintiffs and not under appeal. (Id. at 78.) The buildings were connected by skybridges to one another, to city-owned parking structures, and to two freestanding anchor department stores, also not under appeal. (Id. at 14, 78.) All told, the mall extended over ten parcels, seven of which were not owned by Plaintiffs. (Id. at 21.) The movie DECISION TC-MD 190092G 1 of 26 theater was included in another parcel on the city block northeast of Salem Center North without skybridge access to the mall. (Id. at 21, 78, 83.) The subject consists of the cinema lot and two mall lots containing the North and South buildings’ non-anchor improvements. 1 (Ex A at 2.) The parties have stipulated that the land value for all lots was $30 per square foot. (Id. at 71.)

The mall was built in the 1980s and renovated in 1994 and 2013; on the assessment date it was in average condition. (Ex 1 at 81–85; Ex A at 43, 73.) According to the subject’s December 2017 rent roll, the mall had 173,404 square feet of gross leasable area, excluding the unowned department stores and the cinema. 2 (See Ex 1 at 250–51.) That figure includes 80 square feet for four ATM kiosks and 29,492 square feet for a single major tenant, Ross Dress for Less. (Id. at 245, 250.) Although not broken out separately on the rent roll, Plaintiffs’ appraiser (Booth) allocates 4,150 square feet of the leasable area to food court spaces. (Id. at 81.) Due to its “performance, location, condition, and risk-profile,” the subject’s shopping center was a Class B/B- regional mall. (Id. at 58.)

As of January 1, 2018, the anchor department store spaces were fully occupied by JCPenney, Kohl’s, Macy’s, and Nordstrom. The mall shops had maintained a fairly consistent occupancy rate for several years, ranging from 65 to 75 percent of gross leasable area. (Ex 1 at 138–39; Ex A at 68.) That rate was lower than elsewhere in downtown Salem, where the average retail occupancy rate was 90 to 92 percent. (Ex A at 69.) The subject’s higher-than-average vacancy was partly due to two spaces near the North Mall’s ground floor entrance that had never been leased. (Id. at 75.) The mall’s manager testified that potential tenants have

1 The other mall parcel owned by Plaintiffs is not under appeal. 2 The parties’ appraisers each report a slightly different gross leasable area. Booth reports 172,854 square feet; Tucker reported a net rentable area of 172,584 square feet. (Ex 1 at 81; Ex A at 2.) The difference is immaterial.

DECISION TC-MD 190092G 2 of 26 expressed concerns about those spaces’ physical limitations; restaurant tenants prefer higher ceilings and health care tenants prefer closer parking access.

As of January 1, 2018, the movie theater was fully occupied by Cinebarre. The theater was built in 1988 and renovated in 2009 to a seven-screen dine-in format. (Ex A at 53.) Its amenities include a commercial-sized kitchen, a mezzanine with a bar and common area, and room for electronic games and pool tables. (Id.) It is of average construction quality, similar to the mall shops. (See id. at 53, 73.) Although the rent roll lists the theater’s gross leasable area at 38,000 square feet, Defendant’s appraiser (Tucker) measured the building and found its gross leasable area to be 40,152 square feet. (Id. at 53.) He attributed the discrepancy to the completion of the mezzanine during the 2009 upgrade. B. Trustee’s Sale

The subject was placed into receivership in November 2017 after its prior owners defaulted on their $30 million loan. (Ex 1 at 22; Ex A at 12.) Nationally, the apparel and department store sectors had been facing “disarray and contraction” for several years due to the rise of electronic commerce and big-box retailers. (Ex 1 at 57; see also Ex A at 60–61.) As a result, Class B regional malls faced “uncertain futures due to store closures, department store risk, and a smaller buyer pool of ‘opportunistic’ investors due to the lack of available financing and capital requirements.” (Ex 1 at 73.)

On January 31, 2018, Nordstrom announced it would close its anchor department store at Salem Center in the coming months. (Ex A at 18.) Nordstrom had been consolidating its national portfolio, and “[t]he Salem Center Nordstrom store was their smallest and reported as their lowest performing store remaining in the portfolio at the beginning of 2018.” (Ex 1 at 21.) /// DECISION TC-MD 190092G 3 of 26

A report on the subject appeared in the April 2018 publication Morningstar CMBS Research Mall Monitor – What’s in Store for 2012 Mall Loans? (Ex A at 11.) That publication reported the pending Nordstrom closure was “making it difficult” for the prior owners to refinance their loan. (Id.) The closure of Nordstrom and its mall entrances was predicted to impact foot traffic at the mall. (Id.; see also Ex 1 at 23.) The Morningstar report recommended a modification of the loan and offered the following opinion: “Using a 10.6% capitalization rate, we value the collateral at $27.0 million and estimate an LTV of 114.4%.” (Ex A at 11.)

The Nordstrom at Salem Center ceased operations and closed its doors in April 2018. (Ex 1 at 23; Ex A at 77.)

At some point in this period, a brokerage firm “passed on the opportunity to market and sell Salem Center (including the cinema) because they did not believe it would sell for more than the in-place debt, and likely considerably lower.” (Ex 1 at 22–23.) In reaching that conclusion, the firm relied on a Broker Opinion of Value prepared by Thomas Dobrowski “in 2018” that estimated the subject’s real market value between $18.6 million and $22 million. (Id.)

On August 30, 2018, eight months after the assessment date, Plaintiffs purchased the subject at a trustee’s sale for $27,000,000. (Ex A at 11; cf. Ex 1 at 22.) The interests purchased were leased fee and, to the extent improvements were built on leased ground, leasehold. (Ex A at 8.)

Defendant originally assessed the real market value of the two accounts comprising Salem Center’s non-anchor space at $25,149,270 and the account comprising the theater at $5,262,130. (Ex A at 2.) The Marion County Board of Property Tax Appeals sustained those values. (Compl at 3-5.) On appeal, Plaintiffs request that the court reduce the mall’s real market value to $22,000,000 and the theater’s to $3,600,000. (Ex 1 at 4.) Defendant has reconsidered DECISION TC-MD 190092G 4 of 26 its assessment and now requests that the court increase the roll values to $26,700,000 for the mall and $7,300,000 for the theater. (Ex A at 4; Def’s Trial Memo at 1.)

The parties’ evidence of value is discussed in the analysis below.

II. ANALYSIS

At issue are the real market values of the mall and the movie theater. It is Plaintiffs’ burden to prove those values are lower than the tax roll; it is Defendant’s burden to prove they are higher. See ORS 305.217. 3 A. Highest and Best Use

The parties’ appraisers both conclude the subject’s current use (as of the assessment date) was its highest and best use, although they differ in their forecasts of that current use’s long-term viability. (Ex 1 at 96; Ex A at 77.) Booth recommends “interim use as regional shopping center and adjacent cinema development as currently improved, until feasible redevelopment is warranted, considering the potential to reduce the retail footprint of the mall and add complementary mixed-use considerations.” (Ex 1 at 96.) Tucker recommends the mall continue its current use and “invest in capital improvements to attract creditworthy long-term tenants and lower their average vacancy rates.” (Ex A at 77.)

The appraisers’ differences about the subject’s future highest and best use do not manifest in their valuation evidence. Booth, who favors partial redevelopment after an “interim” period, provided comparables of malls that were not immediately redeveloped. Tucker, who favors capital investment to reduce vacancy in the existing use, concluded the subject’s actual vacancy was its market vacancy. Based on the evidence, the court finds the subject’s highest and best use is adequately described as continuing in its current use.

3 The court’s references to the Oregon Revised Statutes (ORS) are to 2017.

DECISION TC-MD 190092G 5 of 26

Defendant raises another apparent issue as to whether a distinction should be drawn between the mall and the movie theater based on their investment class. Tucker argues the movie theater and mall are investments that appeal to different classes of buyers, with the theater having broader appeal because it is less expensive, less risky, and requires less management. (Ex A at 74.) Booth’s response during testimony was that malls and movie theaters “go together.”

The evidence bears out the contention that malls and movie theaters have different risk profiles: both appraisers concluded the movie theater’s capitalization rate should be lower than the mall’s. However, the evidence does not show that the parties’ differences resulted in differences in their choices of comparables. Booth’s observation that malls and movie theaters go together did not prevent him from selecting a movie theater in downtown Reno without a nearby shopping center as a comparable. (Ex 1 at 207.) Tucker, for his part, included a movie theater adjacent to a mall in Logan, Utah. (Ex A at 83-84.)

The court concludes that the mall and movie theater should be valued separately, but that the appraisers’ different perspectives on the market do not suffice for excluding any particular comparables. B. Mall

In valuing the mall, both appraisers placed primary reliance on an income approach using direct capitalization. (Ex 1 at 117–172; Ex A at 99–110.) Booth also developed a sales comparison approach, which he used as “supporting evidence and as a final check.” (Ex 1 at 99– 116; 173.)

1. Income Approach

Using the direct capitalization method, the parties’ appraisers derived values for the mall by dividing a net operating income by an overall capitalization rate. DECISION TC-MD 190092G 6 of 26

a. Net Operating Income

The parties’ appraisers do not greatly differ in their conclusions of the mall’s net operating income, with Plaintiffs’ appraiser, Booth, actually reaching a higher net operating income ($2,764,427) than Defendant’s appraiser, Tucker ($2,405,125). (Ex 1 at 148; Ex A at 107.)

Booth separately considered the potential income of the in-line shops, food court spaces, ATM kiosks, and the major tenant space. With respect to the in-line shops, he analyzed actual in-place rents according to the varying sizes of their spaces, adjusting all leases to a net basis and separately considering recent leasing to better estimate current market rent. (Ex 1 at 118–21.) Booth further considered retail rent comparables from several other shopping centers in the Willamette Valley and performed a rent-to-sales analysis using average ratios from centers of varying sizes, data from publicly traded companies, and the actual sales of the subject’s in-line shops. (Id. at 121–24.) He performed a similar analysis for the food court spaces and an analysis of major tenant lease comparables for the major tenant space. (Id. at 125–28; 130–31.) Booth determined expenses by comparing the subject’s actual expenses to comparables for numerous specific categories, such as insurance, utilities, repairs and maintenance, janitorial, management fees, et cetera. (Id. at 139–44.) He checked the reasonableness of his conclusions against the overall expenses of shopping center comparables, regional surveys, and publicly traded companies’ reported operating expense ratios. (Id. at 144–49.)

Tucker determined that the rents at the subject mall had been stable for a long time, suggesting actual rents reflected market rents when adjusted for lease structure. (Ex A at 100.) He derived the mall shops’ potential gross income by averaging the total costs of occupancy under the actual leases and cross-checking against typical occupancy costs as reported in an DECISION TC-MD 190092G 7 of 26 appraisal previously prepared by Booth’s firm. (Id. at 99–103.) He accepted the major tenant space’s contract rent as market rent after reviewing recent comparable leases in the Salem area. (Id. at 106.) He also accepted actual expenses (less ground rent) as market. (Id. at 107.)

While both appraisers put weight on the subject’s historical operations, Booth’s analysis provides more detail and includes more cross-checks against market data. It therefore provides more ground for concluding he has abstracted from “the specific circumstances or peculiarities associated with specific owners of property.” Boise Cascade Corp. v. Dept. of Rev., 12 OTR 263, 266 (Or Tax 1991) (stating appraisers’ duty to ignore such circumstances) (quoted in Powell St. I, LLC v. Multnomah Cty. Assessor, 365 Or 245, 445 P3d 297 (2019)). In addition, Booth’s attention to recent leasing more plausibly captures market conditions as of the assessment date. Booth’s conclusion that the mall’s net operating income was $2,764,427 is credible.

b. Capitalization Rate

The appraisers differ widely in their concluded capitalization rates for the mall, with Booth concluding to a 12.50 percent capitalization rate and Tucker concluding to a 9.00 percent capitalization rate. (Ex 1 at 165; Ex A at 109.) The appraisers likewise differ in their methods for extracting a capitalization rate.

Booth derives his capitalization rate from an analysis of comparable sales, market participant interviews, investor surveys, and mortgage-equity analysis. (Ex 1 at 156–65.) Booth’s comparable sales consist of shopping centers across the West, also used in his sales comparison approach. (Id. at 106, 156.) Except for the Flagstaff Mall in Arizona, the centers tend to be considerably larger than the subject, but Booth found them “mostly similar to the subject along a number of lines, including project orientation, tenancy, and so forth.” (Id. at 156.) Nevertheless, Booth admitted that “only a few of the centers are truly viewed as being DECISION TC-MD 190092G 8 of 26 ‘comparable’ to the subject.” (Id.) Of the seven, all but the Flagstaff Mall sale included some form of anchor space. (Id. at 195–202; 112.) Two were sold together in a portfolio. (Id. at 196– 97.) Another was a sale by a loan servicer after a deed in lieu, and another was an uncommonly quick all-cash sale. (Id. at 198–99; 200–01.) Booth ultimately determined the comparable sales supported a range of probable capitalization rates between 12 and 13 percent, “a rate in the middle to lower-end of the comparable sales.” (Id. at 156–57.) That range is consistent with Booth’s interpretation of investor surveys, with the range typical of Class B- malls, and with Booth’s application of mortgage-equity analysis. (Id. at 158–65.) Booth ultimately selected a capitalization rate of 12.50 percent, equal to the rate derived from the sale of the Flagstaff Mall. (Id. at 165, 200.)

Tucker’s list of comparables includes sales of grocery-store anchored neighborhood centers, which he argues resemble the subject because of the subject’s proximity to a population in West Salem underserved by retail, but Tucker did not verify those sales and avers no reliance on them. (Ex A at 98–99, 109.) Instead, he derives a capitalization rate from the Morningstar report and the subject’s 2018 sale, adjusted to fee simple on the basis of market participant interviews reported in an article in the New England Real Estate Journal. (Id. at 107–09.)

The April 2018 Morningstar report recommended a value of $27,000,000 for the entire subject, mall and movie theater together, on the basis of a 10.6 percent capitalization rate. (Ex A at 107.) In Tucker’s view, the subject’s sale for $27,000,000 shortly thereafter confirms 10.6 percent was the “actual 2018 capitalization rate” for the subject’s leasehold interest. (See id. at 108.) According to Tucker, an article by Frederick Richard in the October 19, 2018, issue of the New England Real Estate Journal, Valuation of Leasehold Improvements in a Ground Lease, reported that investors, brokers, and lenders “indicated spreads above leased fee rates ranging DECISION TC-MD 190092G 9 of 26 between 50 and 200 basis points, with one respondent indicating minimal impact on the going-in capitalization rate, depending on the quality of the asset.” (Id. at 107–08.) Because of the ground leases in place, Tucker predicts a range of capitalization rates “1.0 to 2.0 percent lower than for the leasehold, or 8.6 to 9.6% based on the actual 2018 capitalization rate.” (Id. at 108.) Tucker’s concluded rate of 9.0 percent is in line with that range. (Id. at 109.)

As Tucker admitted on cross-examination, the subject’s 2018 sale was not arm’s-length; it was a sale out of receivership at an auction with a single bidder. Evidence indicates that by the end of 2017 apparel and department stores were undergoing a period of disruption due to “the ascent of Amazon and the encroachment of discounters,” a period marked by an increasing number of major chain store closures. (Ex 1 at 57.) Given that disruption and the lack of market-testing, any conclusion based on the subject’s sale price is suspect. That price could be skewed not only by duress on the part of the seller, but by failure on the sole bidder’s part to adapt to a fast-changing retail situation. Furthermore, Tucker’s adjustment for a fee simple valuation based on an article reporting the opinions of market opinions is thinly supported. 4

The inclusion of anchor department store spaces among Booth’s sales comparables decreases confidence in them, and the comparables’ wide range of capitalization rates illustrates that uncertainty, but there appear to be very few comparables without additional nonshop spaces. Even Tucker relied on the sale of the subject, which included a 40,000-square-foot movie theater in addition to mall shops. Booth’s comparable sales included anchor spaces of varying sizes— four included no anchor larger than 62,070 square feet—and his concluded capitalization rate

4 Apart from the general obscurity of citations to market participant interviews, no data are given about the highest and best uses of the properties considered by the market participants. The difference between leasehold and leased fee value would surely loom larger in a property undergoing a highest and best use change, where demolition and construction are contemplated. Here, where the subject has a 40-year ground lease and its highest and best use is its current, income-producing use, Booth’s method of removing ground rent from operating expenses appears better calculated to adjust for the difference.

DECISION TC-MD 190092G 10 of 26 equaled that of the sale without any anchor space.

Of the evidence available, Booth’s comparables 5 to 7 appear the most probative of the mall’s capitalization rate. The subject’s own sale was not arm’s-length, and Tucker’s comparables are unverified. Booth’s comparables 1 to 4 differed significantly from the subject, requiring enormous (95 to 180 percent) adjustments to their sale prices. Comparables 5 to 7 are not without problems, notably that the sales occurred many months before the assessment date in a rapidly changing market and two of them included department store space. Nevertheless, they required comparatively little adjustment.

The average of the capitalization rates of Comparables 5 to 7 is 11.1 percent. That capitalization rate falls at the very bottom of the range assigned to Class B- malls in Booth’s mall grading matrix for the first quarter of 2018, which is where the capitalization rate of a Class B/Bmall would be expected to fall. (Ex 1 at 160.) Accepting a net operating income of $2,764,427, the mall’s indicated value at an 11.1 percent capitalization rate would be $24,904,748.

2. Sales Comparison Approach

Booth identified seven comparable mall sales from Oregon, Washington, Colorado, Utah, and Arizona occurring between August 2016 and March 2018. (Ex 1 at 106.) Except for the Flagstaff Mall, the sales included some anchor space—although typically not the largest department store space. (Id. at 195–202.)

Booth made percentage adjustments to his comparable sales’ prices per square foot based on time of sale, market conditions, location, size, condition, and “economics.” (Ex 1 at 107.) The economics adjustment is meant to account for differences in “achievable rents, tenant sales, occupancy and so forth.” (Id. at 156.) As observed above, the first four comparable sales were quite heavily adjusted; the adjusted price per square foot of the least adjusted (Comparable 1) DECISION TC-MD 190092G 11 of 26 was almost double it actual price; that of the most adjusted (Comparable 2) was a 180 percent increase of its actual price. (Id. at 107.) Comparables 5 to 7 were more lightly adjusted, in the range of 5 to 30 percent, with adjusted prices per square foot ranging from $127.08 to $146.23. (Id.)

Booth ultimately concluded to a price per square foot of $125 under the sales comparison approach. (Ex 1 at 112.) The comparables on which Booth placed most weight included two against which he had applied economics adjustments of 150 percent and 130 percent, as well as significant other adjustments. (Id. at 107, 111–12.)

Tucker sharply criticizes Booth’s inclusion of sales with anchor space as comparables, which he asserts is “misleading.” (Ex A at 98.) As an example, he provides 2020 listing prices for two adjacent properties in downtown Salem—“a 54,451 [square foot] 4-story, multi-tenant retail property with mall shop tenants” listed at $173 per square foot, and a 102,500 square foot former department store listed at $46 per square foot. (Id. at 96–98.) If one were to value the two properties together based on sales at those prices, the indicated value would be $90 per square foot, which would not accurately reflect the value of either the multi-tenant retail or the department store. (Id. at 98.) Because the subject here is mall space without anchors, Tucker would not include any such sales.

Booth himself states that his sales comparison approach has “certain vulnerable characteristics” and that some sales transactions “may not be reflective of current market trends.” (Ex 1 at 173.) Tucker’s criticism of including sales with anchor spaces is particularly relevant with respect to calculating a per-square-foot value. The very large adjustments required for four of the seven sales also diminishes confidence in Booth’s sales comparison approach. /// DECISION TC-MD 190092G 12 of 26

Tucker, for his part, made no attempt to adjust the comparable sales he provided and admitted he had not verified them. The court gives his sales comparison approach conclusion no weight.

3. Reconciliation of Mall Value

The court finds the evidence best supports the value of about $24,904,748 indicated by the income approach. That would be a per-square-foot value of $143.62. While the results of the sales comparison approach are inconclusive, the per-square-foot value concluded under the income approach is similar to the adjusted values of Booth’s two least-adjusted comparable sales, Comparable 6 at $139.71 and Comparable 7 at $146.23. (Ex 1 at 107.)

On the available evidence, the court finds the mall’s real market value was $24,900,000 on the assessment date. C. Movie Theater

Both appraisers placed primary reliance on the income approach in valuing the movie theater. (Ex 1 at 173–74; Ex A at 93–94.) Each also developed a sales comparison approach, and Tucker developed the cost approach as well.

1. Income Approach

The appraisers used direct capitalization in their valuations of the movie theater. They differed significantly in their conclusions of market rent and overall capitalization rate.

a. Net Operating Income

Booth estimated the subject movie theater’s market rent at $11.00 per square foot, while Tucker estimated it at $15.00 per square foot. (Ex 1 at 167–68; Ex A at 90–92.) Both appraisers relied primarily on rent comparables, with some weight given to the subject’s lease in place. /// DECISION TC-MD 190092G 13 of 26

Booth supplied a list of nine movie theater rent comparables with brief comments. (Ex 1 at 168.) One comparable—a 2014 lease in South Jordan, Utah—was the lease of a very large building (125,899 square feet), while the other eight comparables were leases of buildings ranging from 26,400 to 68,000 square feet. (Id.) The average initial rent per square foot of all nine comparables was $11.21. (Id.)

Booth reported the in-place lease rate for the subject movie theater was $10.00 per square foot—a point that Tucker controverts. (Ex 1 at 167.) The November 2017 rent roll shows the subject’s lease had a start date of November 6, 2009, with an annual base rent of $10.00 per square foot, or $380,000 for 38,000 square feet. (Id. at 238.) That same rent roll indicates “percent/overage rent” of 7.00 percent on sales over $5.5 million. (Id.) The subject’s sales at the time of the assessment date were $4.1 or $4.2 million, implying that no percent/overage rent was due. (Id. at 167; Ex A at 92.)

Booth estimated a market rent of $11.00 per square foot after considering the rent comparables’ range of $8.00 to $16.00, their average of $11.21, the in-place lease of $10.00, and the subject’s location, quality, and achieved sales. (Ex 1 at 168.)

Tucker suppled a list of ten movie theater rent comparables that he had previously received from Booth’s firm. (Ex A at 90–91.) Five of the sales on Tucker’s list appeared to be the same as those on Booth’s list, but there were some discrepancies among those five; e.g., the South Jordan comparable was stated at 182,000 square feet on Tucker’s list. (Compare id. at 91 with Ex 1 at 168.) The comparables closest to the assessment date from Booth’s list were not present on Tucker’s list. (See id.) Tucker analyzed his list as follows: “Excluding the lease renewals, and the lease for 182,000 SF, the three remaining leases indicate a market lease rate range of $14.15 to $17.76 per square foot. Comp 4 at $17.76/SF is the lease of a dine-in DECISION TC-MD 190092G 14 of 26 theater.” 5 (Ex A at 92.) The average initial rent per square foot of the three comparables retained by Tucker (Comparables 2, 4, and 5) was $16.08. (See id. at 91.)

Contradicting Booth, Tucker reported the in-place lease rate for the subject was $13.90 per square foot under the terms of the lease. (Ex A at 92.) The subject’s tenant’s August 2018 lease renewal letter references “that certain Lease dated September 19, 1988, as amended.” (Id. at 125.) A single page of what appears to be that original lease was provided, containing Paragraph 5 (“Base Rental”), a portion of the preceding paragraph, and a portion of the succeeding Paragraph 6 (“Percentage Rental”). (Id. at 126.) Paragraph 5 requires a base rent of $40,500 per month, “subject to adjustment under Section 4.” (Id.) The portion of Paragraph 6 provided appears to require percentage rent of 10 percent of general sales exceeding $3,742,000 and 6 percent of specialty sales exceeding $1,117,800, “subject to adjustment under Section 4 with respect to Renewal Terms.” (Id.) Under those terms, Tucker computed an in-place lease amounting $13.90 per square foot.

Tucker estimated a market rent of $15.00 per square foot—halfway between the subject’s actual rent and the average of the comparable rents—“[b]ased on the fairly narrow range of comparable rents, and putting some emphasis on the actual lease with percentage rent[.]” (Ex A at 92.) Tucker stated that a market rent of 14 percent of gross sales was “in the reasonable range.” (Id.)

In the court’s view, Booth’s list of rent comparables is more probative than Tucker’s. Booth personally generated his list, whereas Tucker was relaying another appraiser’s work, and Booth’s list included leases closer to the assessment date. In analyzing his own list, Tucker excluded the largest comparable and all comparables identified as lease extensions, with the

5 The initial rent per square foot of Defendant’s movie theater rent comparable 4 is $17.78. (Ex A at 91.)

DECISION TC-MD 190092G 15 of 26 result that he based his conclusion on three comparables. If Tucker’s analysis were applied to Booth’s list, six comparables would be left—Comparables 2, 3, 4, 5, 7, and 9. The average lease rate of those six is $11.22 per square foot—only a penny more than the average of the entire list—and the range is $8.00 to $14.15. (See Ex 1 at 168.)

Tucker argues that an actual lease is better evidence of a rental rate than a rent roll. That may well be, but no complete lease is in evidence. The single page available does not show who its signatories are and does not contain the whole paragraph dealing with percentage rent. The portion it does contain references another section allowing for adjustment of rents during renewal terms. Considering the 30-year interval since the original lease was written, during which the subject was converted to a dine-in theater, it would indeed be surprising if lease terms relating to sales of concessions were not revisited. On the balance of the evidence, the rent stated in the rent roll is more likely to reflect the subject’s actual rent.

Tucker’s discovery of additional mezzanine space was not contradicted, and the court accepts that the subject’s gross leasable area was 40,152 square feet, rather than the 38,000 square feet indicated on the rent roll. Because the subject’s actual per-square-foot rent is calculated from the rent roll, a larger gross leasable area than indicated on the rent roll would imply a lower rent rate per square foot—$9.50 instead of $10.00. (See Ex 1 at 238.) Considering that both appraisers gave weight to the subject’s actual rent, a lower per-square-foot rent should presumably lower the market rent estimate.

The court accepts Booth’s analysis of market rent and Tucker’s observation of the subject’s gross leasable area. Slightly reducing Booth’s concluded market rent based on the above analysis, the court finds a market rent of $10.75 per square foot. The resulting potential /// DECISION TC-MD 190092G 16 of 26 rental income for the subject is $431,634, a bit higher than the $418,000 calculated by Booth. (See Ex 1 at 170.)

In deriving net operating income, Booth and Tucker accounted for losses and expenses differently. Booth included market-derived payments totaling $141,619 for repairs, insurance, and tax as reimbursement income within the subject’s potential gross income, calculated effective gross income by deducting a five percent vacancy and collection loss, and then deducted the reimbursement income items and a three-percent management fee to determine net operating income. (Ex 1 at 170.) Tucker allowed no vacancy and collection loss (“there is no vacancy”), deducting only a three percent expense “for replacement reserves and a small amount of owner’s expenses.” (Ex A at 92.)

Booth’s method of first including and then removing tenant-paid items as reimbursement income accounts for revenue lost due to vacancy: the owner must pay those expenses if there is no tenant. Although the subject theater has not yet been vacant, the evidence indicates movie theaters can and do go through periods between tenancies. (See Ex D.) It is reasonable to account for such vacancies, even in a stabilized space typically leased for a long time. The court accepts Booth’s five-percent vacancy rate and his method of deriving net operating income. The three-percent expense (either for management or replacement reserves) is also accepted. Adding $141,619 in reimbursements to $431,634 in potential rent, then deducting 5 percent of the total for vacancy losses, yields an effective gross income of $544,590. Deducting the reimbursements and a 3-percent management fee yields a net operating income of $386,634, or $9.63 per square foot. /// /// DECISION TC-MD 190092G 17 of 26

b. Capitalization Rate

The appraisers derived capitalization rates for the subject movie theater from their respective lists of unadjusted comparable sales prepared for the sales comparison approach; Tucker also consulted a market survey. (Ex 1 at 170–72; Ex A at 92–93.) Although the comparables were not adjusted, both appraisers provided analysis of the relative market values of the comparables compared to the subject. (Ex 1 at 114–15; Ex A at 86, 88.)

Booth found his Comparables 2 and 3 most probative of the subject’s capitalization rate, setting an initial range of 9.28 to 10.80 percent. (See Ex 1 at 171–72.) He considered the overall median (10.04 percent) and average (11.78 percent) of his entire list of comparables, as well as the fact that the subject sold “without additional land and parking areas typically included with other cinema sales,” and concluded to a capitalization rate of 10.50 percent for the subject. (Id.)

Tucker determined the average capitalization rate of all his comparables was 8.18 percent, concluding that the subject’s capitalization rate was 8.0 percent on January 1, 2018. (Ex A at 92.) Tucker noted that “[a]ccording to CoStar, the cap rates for single-tenant movie theater sales have risen 100 basis points in the last five years to a current average of 7.50 percent, with the majority of the increase coming 2017 and 2018[.]”

Surveying the evidence, the court finds Booth’s Comparables 3 and 5 most probative of the subject’s capitalization rate. Booth’s Comparable 3, located in an Atlanta suburb and selling in November 2018 with a capitalization rate of 9.28 percent, resembled the subject in relying on a neighboring shopping center for parking. (Ex 1 at 114-15.) Booth’s Comparable 5, a movie theater in downtown Reno, similarly lacks on-site parking. The court notes the difference between its listing capitalization rate (8.00 percent) and December 2016 sale capitalization rate (8.81 percent) illustrates the likelihood of a gap between data derived from listings and from DECISION TC-MD 190092G 18 of 26 sales. (Id. at 207.) Although Booth had put more weight on Comparable 2, that property underwent “major renovations” immediately after sale, suggesting that its capitalization rate of 10.80 percent is a high indicator. (Id. at 204; Ex B at 7.)

Among Tucker’s comparables, his Comparables 1 and 2 sold in April 2015 and June 2014 with capitalization rates of 8.00 and 7.00 percent, respectively. (Ex A at 83, 86.) The evidence indicates capitalization rates for movie theaters were rising over the years after those sales. (Id. at 92.) Comparables 1 and 2 are probably low indicators of the subject’s capitalization rate on the assessment date.

Other comparables are less indicative. Tucker’s Comparable 5 was a listing, not a sale. (Ex A at 83, 85.) Booth’s Comparable 1 and Tucker’s Comparables 3 and 4 were components of portfolio sales, raising questions about allocating value among several properties sold. (Ex 1 at 203; Ex A at 84–85.) Booth’s Comparables 4 and 6 had outlying capitalization rates—19.54 percent and 14.74 percent. (Ex 1 at 206, 208.) Booth placed less weight on those sales, noting that the latter was in “a more rural location.” (Id. at 115.)

All told, the evidence supports a capitalization rate of 9.00 percent for the subject as of January 1, 2018. That amount approximates the rates of Booth’s Comparables 3 and 5 and, as expected, exceeds the rates implied by Tucker’s Comparables 1 and 2. The court finds a capitalization rate of 9.00 percent.

With a capitalization rate of 9.00 percent and a net operating income of $386,634, the value of the movie theater indicated by the income approach is $4,295,933.

2. Sales Comparison Approach

Booth identified eight sales of movie theaters around the country, including two in Oregon. (Ex 1 at 114.) Tucker identified four sales in the western United States, including one DECISION TC-MD 190092G 19 of 26 in Oregon and one in Vancouver, Washington—the latter the only dine-in theater among all the comparables. (Ex A at 83.) Tucker also included a 2017 listing for a theater in Indiana. (Id.)

Although the listed comparables varied widely in size (23,960 to 74,170 square feet), date of sale (June 2014 to June 2019), and number of screens (5 to 17), neither appraiser made any adjustments. (Ex 1 at 114, 204-210; Ex A at 83.) Booth stated: “It is difficult to relate the subject to comparable properties that are in such widely divergent markets with different cash flow characteristics.” (Ex 1 at 115.) Given those difficulties, the appraisers took different paths: Booth performed a qualitative comparison of the subject to the comparables, while Tucker applied linear regression analysis.

a. Qualitative Comparison

Booth identified relevant characteristics of the comparables as superior, similar, or inferior to the subject, developing a range of value for the subject on both a per-square-foot and a per-screen basis. (Ex 1 at 113–16.) He gave most weight to his comparables 1, 3, and 8. (Id. at 115.)

Booth’s Comparable 1 was similar in size and location to the subject and superior in condition. (Ex 1 at 115.) It sold as part of a portfolio, with a sale price listed at $115.29 per square foot, equal to $448,354 per screen. (See id. at 114–15; 203.) Booth estimated the subject’s value would be below Comparable 1’s. (Id.) Comparable 8 was similar to the subject in condition and quality, somewhat inferior in location, and relatively old in that it closed in July 2015. (Id. at 114-15.) It sold for $74.55 per square foot, equal to $240,000 per screen. (See id. at 114, 210.) Booth estimated the subject’s value would be above Comparable 8’s. (Id. at 115.) Comparable 3 was similar to the subject in lot size and parking arrangement with a neighboring /// DECISION TC-MD 190092G 20 of 26 shopping center; it sold for $86.23 per square foot, equal to $212,500 per screen. (See id. at 114–15, 205.) Booth did not specifically relate its value to the subject’s.

Booth concluded the subject’s value would lie within the range of $90 to $100 per square foot. (Ex 1 at 115.) At $95 per square foot, the indicated value would be $3,610,000 “prior to adjustments for underlying land.” (Id. at 115-16.) That would be equivalent to a real market value of $515,714 per screen—a value exceeding the range of Booth’s comparable sales’ per-screen values. (Id. at 115–16.)

Booth’s qualitative comparison is a reasonable way to address the limited data available. However, its inherent uncertainty is increased by the lack of adjustments and by Booth’s heavy reliance on a portfolio sale-leaseback. Furthermore, Booth’s concluded value falls outside of his comparables’ per-screen value range, removing one possible additional check on his conclusion. Very little weight can be placed on Booth’s sales comparison approach.

b. Linear Regression

Tucker’s sales comparison approach used linear regression to establish trendlines correlating two sets of data drawn from his sales and listing comparables.

Tucker first found what he determined to be a strong correlation between price per screen and area per screen based on a 0.8763 coefficient of determination. (Ex A at 86–87.) The court’s plot of Tucker’s data is shown in Table 1 below. 6

6 The court’s plot of the data shows a negligible rounding difference.

DECISION TC-MD 190092G 21 of 26

Table 1. Square Feet per Screen v. Price per Screen, Tucker’s Comparables

$1,800,000.00 $1,600,000.00 Comp 1 $1,400,000.00 y = 260.65x - 292726 $1,200,000.00 R² = 0.8762 Comp 4 $1,000,000.00 Comp 5 $800,000.00 Comp 2 $600,000.00 Comp 3 $400,000.00 $200,000.00 $- 2,000 3,000 4,000 5,000 6,000 7,000 8,000

Based on Tucker’s trendline, the subject’s 5,736 square feet per screen yields a per-screen value of $1,202,388, with a standard error of $156,150. Tucker concluded that “a value per screen between $1,100,000 and $1,200,000 is indicated.” (Id. at 88.)

Tucker next applied linear regression to correlate his comparables’ net operating income per square foot with price per square foot. (Ex A at 89.) Tucker’s analysis excluded Comparable 4 because its net operating income was unknown. When charted, Tucker’s data appear as shown in Table 2.

Table 2. Net Operating Income per Square Foot vs. Price per Square Foot

$250.00 y = 9.5552x + 42.047 Tucker 1Tucker 5 $200.00 R² = 0.6856

Tucker 2 $150.00 Tucker 3

$100.00 $10.00 $11.00 $12.00 $13.00 $14.00 $15.00 $16.00 $17.00 $18.00 $19.00

Using a value of $15 per square foot (the market rent he had determined in his income approach) Tucker determined an indicated value of $190 per square foot. 7 (Id.)

Tucker calculated a value of $188.00 per square foot using a slope of 10.55 and a y-intercept of 29.25. 7

(Ex A at 89.) The court’s analysis of Tucker’s data yields a slope of 9.5552 and a y-intercept of 42.047. Under those parameters, the indicated value using Tucker’s selected net operating income would be $182.54 per square

DECISION TC-MD 190092G 22 of 26

Tucker’s total indicated values were $8,400,000 from his area-per-screen analysis and $7,626,030 from his net-operating-income-per-square-foot analysis. He reconciled those results strongly in favor of the net operating income method, concluding to a value of $7,700,000 under the sales approach.

This court has previously observed that where a linear regression analysis based on unadjusted market data is the only evidence supporting a market comparison approach, “the reliability of the indication of value is questionable.” Gangle v. Dept. of Rev., 13 OTR 343, 349 (1995). In this case, that is doubly so because there are so few data points.

Tucker’s Comparable 1 is an outlier along the x-axis in his area-per-screen analysis; no other comparables approach its size per screen, even among Booth’s comparables. Its disproportionate influence on the coefficient of determination is evident when it is omitted, as shown by comparing Table 3 below with Table 1 above.

Table 3. Square Feet per Screen v. Price per Screen, Outlier Omitted

$1,400,000.00 $1,200,000.00 Comp 4 $1,000,000.00 y = 369.9x - 760893 R² = 0.6021 Comp 5 $800,000.00 Comp 2 $600,000.00 Comp 3

$400,000.00 $200,000.00 $- 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500

Without Tucker’s Comparable 1, the coefficient of determination (“R2”) drops by a third—a dramatic illustration of how dependent the model is on a single data point. With such limited data, it is impossible to confidently predict the subject’s per-screen value from Tucker’s model. foot. Without access to Tucker’s underlying calculations, the court is unable to reconcile the discrepancy. DECISION TC-MD 190092G 23 of 26

Tucker’s net operating income analysis relies on fewer data points and has a lower coefficient of determination than his area-per-screen analysis; on its face, it is even less reliable. 8 Furthermore, the court’s finding of the subject’s net operating income differs significantly from the $15-per-square-foot net operating income used by Tucker here. 9

A sales comparison approach exclusively relying on even a well-prepared linear regression analysis of unadjusted comparables would be questionable as an indicator of value. See Gangle, 13 Or Tax at 343. Given the dearth of data here, as well as unexplained discrepancies in the analysis, the court cannot place reliance on Tucker’s sales comparison approach.

3. Cost Approach

Tucker prepared a cost approach using the Marshall Valuation Service. (Ex A at 79–81.) He determined the subject’s base cost of construction as of 2019, together with the Local Cost Multiplier appropriate to Salem, was equal to $6,896,932. (Id. at 79–80.) Although the subject was built in 1988, Tucker estimated its effective age at 20 years because of the 2009 renovation. (Id. at 80.) He therefore subtracted 30 percent of his calculated cost of construction based on the Marshall Swift estimate of physical depreciation. (Id.) Having thus calculated the subject’s value as of 2019, he subtracted an additional 5 percent to adjust to the assessment date. (Id.) His concluded value for the improvements was $5,045,106. (Id.) He then added in the stipulated value of the land, concluding under the cost approach to an indicated value of $6,340,000. (Id. at 81.)

8 The evidence does not show why Tucker placed more weight on it than on the per-screen analysis in his reconciliation. 9 The evidence does not show why Tucker used the $15 per square foot had concluded for the subject’s market rent rather than the $14.55 per square foot he had concluded for its net operating income. (Cf. Ex A at 92.)

DECISION TC-MD 190092G 24 of 26

Tucker opined that his cost approach indicated a value too low, apparently because it did not reflect the value the subject receives from offsite parking at the mall. He noted that movie theaters typically have large parking lots, whereas the subject had no need for one.

Tucker’s reasoning on the effect of offsite parking on the cost approach is not entirely clear to the court, and there may be a misunderstanding. To the court, it appears that the value of the subject’s offsite parking is accounted for in a stipulated land value exceeding the land value of movie theaters with large parking lots. Contrary to Tucker’s opinion, the cost approach appears to serve as a check against excessive valuation: If the subject could have been constructed brand new in 2019 for $6.9 million, one would expect its value to have been considerably less 30 years after it was built and 9 years after it was renovated. Nevertheless, considering the age of the subject property, the court agrees with Tucker that the cost approach is of limited value.

4. Reconciliation

Given the available evidence, the court finds that the income approach yields the best supported evidence of value. The movie theater’s real market value as of January 1, 2018, was $4,300,000. /// /// /// /// /// /// /// DECISION TC-MD 190092G 25 of 26

III. CONCLUSION

The evidence indicates that both the mall and the movie theater are overvalued on the 2018–19 tax roll. Now, therefore,

IT IS THE DECISION OF THIS COURT that Salem Center’s real market value as of January 1, 2018, was $24,900,000, exclusive of the four anchor department store spaces.

IT IS FURTHER DECIDED that the subject movie theater’s real market value as of January 1, 2018, was $4,300,000.

Dated this ____ day of September 2021.

POUL F. LUNDGREN MAGISTRATE

If you want to appeal this Decision, file a complaint in the Regular Division of the Oregon Tax Court, by mailing to: 1163 State Street, Salem, OR 97301-2563; or by hand delivery to: Fourth Floor, 1241 State Street, Salem, OR.

Your complaint must be submitted within 60 days after the date of this Decision or this Decision cannot be changed. TCR-MD 19 B.

Some appeal deadlines were extended in response to the Covid-19 emergency. Additional information is available at https://www.courts.oregon.gov/courts/tax

This document was signed by Magistrate Poul F. Lundgren and entered on September 29, 2021.

DECISION TC-MD 190092G 26 of 26