GLC-South Hillsboro, LLC v. Washington County Assessor

Opinion

IN THE OREGON TAX COURT

MAGISTRATE DIVISION

Property Tax

GLC-SOUTH HILLSBORO, LLC, )

)

Plaintiff, ) TC-MD 180141R )

v. )

)

WASHINGTON COUNTY ASSESSOR, )

)

Defendant. ) DECISION

Plaintiff appealed a Washington County board of property tax appeals (BOPTA) order, mailed March 12, 2018, for the 2017-18 tax year. A trial was held in the courtroom of the Oregon Tax Court. Alex Robinson, of CKR Law Group P.C., appeared on behalf of Plaintiff. David Brentlinger (Brentlinger) and Matthew Call (Call) testified on behalf of Plaintiff. Jason Bush, Assistant County Council, appeared on behalf of Defendant. Sean Morrison (Morrison) and Kathryn Vai (Vai) testified on behalf of Defendant. Plaintiff’s exhibits 1 to 9 were received into evidence without objection. Defendant’s exhibits A, G, H, I, J and M were received into evidence without objection. Defendant’s exhibit L was received into evidence, for rebuttal purposes only, over Plaintiff’s objection.

I. STATEMENT OF FACTS

This appeal involves a select portion of the largest master-planned community in Washington County history. Plaintiff purchased approximately 1,400 acres of vacant land between Tualatin-Valley Highway and Farmington Road, for $9 million in 2000 in a then unincorporated area just south of the city of Hillsboro (the City). Part of that tract, known as “Reed’s Crossing,” was brought within the urban growth boundary with some other land in 2014 in what Defendant refers to as “the grand bargain” to create “South Hillsboro.” The tract was

DECISION TC-MD 180141R 1 annexed into the City in 2015 and planned development was approved in late 2016. Plaintiff plans to develop the land as part of a master-planned community comprised of single-family homes, multi-family homes, and commercial units. The parties agreed that for purposes of this appeal the issues to be decided by the court are confined to the real market value of approximately 140.2 acres out of a 240.63 tax lot for the 2017-18 tax year, which contains a mix of residential and commercial zoning. Reed’s Crossing is bounded on the west by Southeast 29th and on the east by a BPA power line; on the north by Tualatin-Valley highway, and on the south by Blanton road. The parties agreed that the net developable area of the subject property is 107.72 acres. It is comprised of: 21.32 acres of Single-Family Residential 4.5 (SFR-4.5); 5.94 acres of Single-Family Residential 6 (SFR-6); 8.91 acres of Multi-Family Residential 1 (MFR- 1); 18.28 acres of Multi-Family Residential 2 (MFR-2); 7.84 acres of Multi-Family Residential 3 (MFR-3); and 45.43 acres of Mixed Use Village Town Center (MU-VTC). (Ex 1 at 29.) A. Plaintiff’s Evidence of Value Plaintiff is a joint venture LLC comprised of two primary investors with Newland Communities (Newland) as the general and managing partner. Brentlinger testified he is a vice- president of Newland with general management responsibility for Reed’s Crossing. Brentlinger testified he has a 30-year background in real estate having worked with home builders, master planned communities, lenders, and equity investors. He testified that master planned development is much different than residential subdivisions because the land lacks “entitlements” i.e.—the ability to immediately build upon the land. Brentlinger testified that a rough estimate of the scope of the whole development includes about 2,800 single-family homes, 1,250 apartments/condominiums, parks, trails, and retail space. ///

DECISION TC-MD 180141R 2

Brentlinger testified that as of January 1, 2017, Plaintiff needed to complete the “Gateway Project”—a majority of which included a railroad crossing adjacent to Tualatin Valley Highway and an extension to Cornelius Pass road. Minor projects included the construction of Blanton Street. He estimated that the Gateway Project alone would cost $30 million, not including the necessary storm drainage system. He testified that any potential buyer would have to incur those costs to make the site developable. The costs involved were greater than typical land development and represented risk to potential investors. Brentlinger noted that as of the assessment date a final plat had not been approved and the tract had not been divided into individual lots that could be sold and built upon. Brentlinger testified that because of the size of the tract, the amount of investment required, the long period it would take to generate cash flow, and the risks involved with receiving entitlements; market participants would have used a Discounted Cash Flow (DCF) analysis to evaluate the subject property’s value. In discussing the risks of a potential buyer, he stated that there are conditions of approval that have to go through public hearings which can alter the end use of the property and its value.

1. Plaintiff’s Sales Comparison Approach In October 2018, Plaintiff sold 15.16 acres of MU-VTC zones land located in the northwest corner of Reed’s Crossing to Nash Holland for $5,650,000 ($373,184 per acre). Brentlinger testified that Mr. Holland is an investor in Plaintiff and in Newland. He testified that he believed the sale represented an arm’s-length transaction because the sale was “consistent with other offers we’ve had for multi-family land.” Brentlinger testified on that Plaintiff sold the local school district a 40-acre parcel without a specific location, but the sale was a special deal that involved a lot line adjustment and no other purchaser could do that. ///

DECISION TC-MD 180141R 3

Call testified he has been a commercial real estate appraiser in Oregon and Washington since 1999, a certified general appraiser in Oregon since 2003, and is a member of the Appraisal Institute (MAI). Call performed a physical inspection of the subject property in March 2018 and prepared a retrospective appraisal as of January 1, 2017. (Ex 1.) Call testified that the entire parcel being developed is about 1,400 acres brought into the urban growth boundary and is being developed with three main properties: Reed’s Crossing, Rosedale Park, and Butternut Creek. The subject property is the northwest portion of Reed’s Crossing.

Call explained that as of January 1, 2017, the subject property had been annexed into Hillsboro, had zoning in place, and there was an agreement allowing completion of infrastructure, but no master plan for actual developments had been approved. At the end of 2016, a Local Improvement District (LID) was approved by the City. Call determined the highest and best use of the subject property was as a large master-planned community with developed single-family, multi-family, and commercial/mixed-use development. He estimated that vertical construction would be possible within 18 to 24 months from the date of assessment. Call also concluded that the significant holding period reduces the value of the subject property as of the assessment date.

Call considered the sales comparison and income approaches but dismissed the cost approach as inappropriate for raw land. Call considered a variation of the income approach, the DCF analysis, as superior because of the unique size of the property and because, in his experience, that is how market participants would determine its value. Call’s appraisal report describes the DCF analysis as “where the gross revenue from future developable land sales is estimated (using the values concluded in the Sales Comparison Approach), development and sales costs are deducted, and the anticipated future cash flows are discounted to a present value

DECISION TC-MD 180141R 4 at an appropriate rate to reflect the as-is value of the property as of January 1, 2017.” (Ex 1 at 10.)

For the sales comparison approach to value, Call used three general land-use categories:

single family residence (SFR), multi-family residence (MFR), and commercial/mixed use village town center (MU-VTC). Call considered SFR and MFR values based on a price per acre basis and MU-VTC value based on price per square foot. Call could not find sales of large tracts similar to the subject property in the Portland Metropolitan area, so he used lots which had entitlements and infrastructure and made qualitative adjustments. (See Ex 1 at 44-60.)

a. SFR

Call selected four comparable properties in Washington County and in the Portland area with similar density. (Ex 1 at 45-48.) Comparable #1 is a 74.29-acre lot in Cornelius which sold for $208,642 per acre in September 2016 and brought within the urban growth boundary at the same time as the subject property. Call rated this property as inferior to the subject property due to its location and condition. Comparable #2 is a 57.71-acre lot in the South Cooper Mountain area that sold for $410,918 per acre in May 2018 but had been under contract since 2015. Call considered this location superior due to its location and a high indicator of value and adjusted the price to $436,161 per acre. Comparable #3 is a 19.79-acre lot in Hillsboro which sold for $400,000 per acre in January 2017. Call viewed this sale as “a reasonable to slightly low indicator” of value because the transaction agreement was made in 2016 even though it closed in 2017. Comparable #4 is a 27.95-acre lot in Beaverton which sold for $522,081 per acre in June 2018. Call viewed the location and density, including apartment units, superior to the subject property making this sale a high indicator of value for the SFR category. Call found his comparables indicated a range of $208,642 to $522,081 per acre for SFR sites and concluded that

DECISION TC-MD 180141R 5 the value of the subject property’s SFR lots (SFR 4.5 and SFR 6) was $425,000 per acre based on bracketing Comparable #3 with Comparable #2.

b. MFR

Call found five comparable sales with a range of $373,184 to $723,445 per acre for MFR sites between 3.13 acres and 22.44 net acres. (Ex 1 at 53.) Comparable #1 is a 22.44 net acre site in South Cooper Mountain that sold for in June 2018 for $522,081 per acre. Call found the location slightly superior to the subject property. Comparable #2 is a 10.45 net acres site, adjacent to Comparable #1, that sold in April 2016 for $723,445 per acre. Call found the site had a superior location and an established market area and thus it was a high indicator of value for the subject property. Comparable #3 is a 7.37 net acre site in Happy Valley that sold in August 2015 for $512,212 per acre. The site is part of a developing area in Happy Valley with similar density and a planned commercial area. Call determined that the lower cost of development makes this site a slightly high indicator of value for the subject property. Comparable #4 is a 3.13 net acre site in Beaverton that sold in April 2016 for $552,716 per acre. Call found the density and locations reasonably similar to the subject property with slightly higher density, but its smaller size makes it a high indicator of value for the Subject Property. Comparable #5 is a 15.14 net acre site in Hillsboro that sold in October 2018, but was negotiated in late 2016 and early 2017, for $373,184 per acre. This site is multi-zoned MFR-1, MFR-4 and MU-VTC. Call found that because of the higher costs of the MU-VTC zone, this comparable is a slightly low indicator of value for the subject property. Call determined that Comparable #5 on low end and #3 on the high end bracketed the subject property, leading to his concluded value at $450,000 per acre for the MFR zone. ///

DECISION TC-MD 180141R 6

c. MU-VTC

Call selected four comparable sales for the MU-VTC zone. He used a price per square foot to analyze the tracts and then converted the price into acres. Call noted that expenses for development of this zone type significantly affect the price. Comparable #1 was 259,182 square feet and sold for $3.8 million in Feb 2015, equating to $15 per square feet. The site was purchased by Nike and not for development and thus Call determined this comparable was a low indicator of value. Comparable #2 was 217,800 square feet and sold for $3.6 million in April 2015, equating to $17 per square foot. Comparable #3 was 197,762 square feet and sold for $3.956 million in May 2016, equating to a price of $20 per square foot. The site was an industrial zone site without retail appeal, and thus inferior to the subject property. Call believed the location was slightly superior and the small size put an upward pressure on the price. Call found it a slightly high indicator of value for the subject property. Comparable #4 was a 158,994 square foot site and sold in March 2017, for $4.057 million equating to $26 per square foot. Call found the comparables indicated a range of $15 to $26 per square foot and concluded that because of its large size the subject property would have to be sold in phases thus reducing its value on balance with the comparables. Additionally, Call found the high LID costs associated with the subject property also puts downward pressure on the value. Call’s concluded value for this zone is $16 per square foot which translates to approximately $700,000 per acre.

d. Sales Comparison Approach Value Totals Call multiplied the acreage for each of the three zones identified above by the price per acre resulting in a total of $59,150,000. (Ex 1 at 67.) From that figure he deducted the allocated infrastructure costs of $17,920,771, resulting in a net value of $41,230,000. Call testified he did not make any adjustments for reimbursement amounts for infrastructure LID or TDT because

DECISION TC-MD 180141R 7 that would go into a DCF analysis, which he performed separately. Call testified that his sales comparison approach was not as accurate as the DCF approach because it does not address the significant investment required, the timing of development, and the revenue spread out over time.

e. Butternut Creek comparison As alternate to the sales comparison approach, Call analyzed a large related development property known as Butternut Creek located just south of the subject property. This property includes SFR, MFR and MU-VTC zoning with approximately 1,252 units planned for the 176- acre site. In 2016, a builder entered a purchase contract for Butternut Creek broken into nine parts, with one part per year. The initial transaction occurred in September 2017, with the sale of 44.97 acres for $5,863,250. The second transaction in August 2018 involved 20.77 acres for $4,095,466. The remaining sales were tied to the county real estate index but capped at five percent. (Ex 1 at 69.) Based on the net developable property, Call determined the net sales prices, in the range of $235,000 to $289,000 per acre represented the “absolute top end of a supportable market value for the entire property.” (Id.) Using that range, Call determined that the price of the subject property, using Butternut Creek as the model, would be in the range of $25,310,000 to $31,130,000.

2. Discounted Cash Flow Analysis The DCF analysis is an income approach that starts with a sales comparison approach and modifies it to “simulate the anticipated cash flows during the sellout period” and discounting the yield “to reflect the time value of money and required profit.” (Ex 1 at 61.) Call testified that this approach more accurately represents how market participants would evaluate the subject property. Call grouped the value of the residential zones together from his comparison approach

DECISION TC-MD 180141R 8 values and found a blended rate for that category at $439,059 per acre allocated to 62.29 acres. To that value he added the MU-VTC zones at $700,000 per acre allocated to 45.43 acres. (Id.) He assumed appreciation at four percent which he felt was in line with the Butternut Creek sale which was capped at five percent. (Id.)

Call estimated a three-phase sellout period with the first phase beginning in 2018 and the second and third phase beginning in 2020 and 2022. Call calculated the expenses related to completion of the Gateway infrastructure which include the rail crossings at Cornelius Pass Road at Tualatin Valley Highway, construction of Cornelius Pass Road and Blanton Street, and off-site transportation improvement costs. (Id. at 62.) Because this appeal was only for 140 acres out of a 240-acre tax lot, Call allocated the expenses to the subject property at $6,796,612 for 2017 and $11,124,159 for 2018. (Id.) Call used estimates from Plaintiff in calculating TDT credits at 50 percent of the road costs or $9,770,043. (Id. at 63.) Call asserted that TDT credits are realized at the time of vertical construction, but that timeline was uncertain, so he amortized the amounts over five years or $1,954,009 per year (total $9,770,043). (Id.) Call also anticipated a $10.5 million future payment for LID and estimated the payment to come in 2023. (Id.) Allocating that amount to only the subject property, Call included a credit for LID at $6,489,570 in 2022 which is the last year of the hypothetical sell-out period. (Id.)

Call described the discount rate as the amount a potential buyer would need to account for “entrepreneurial profits and overhead, and the cost of capital.” (Id.) Call used the first quarter 2017 Developer Survey published by Realty Rates.com for residential/mixed-use subdivisions over 500 units in the Pacific Northwest which ranged from 14.32 to 32.13 percent and averaging 21.54 percent. (Id.) Call also considered Brentlinger’s discount rate for similar projects ranging from 22 to 25 percent. (Id.) Call considered the discount rate in D.R. Horton’s

DECISION TC-MD 180141R 9 2016 annual report which ranged from 10 to 14 percent and homebuilder Lennar’s rate at 20 percent. (Id.) Call noted that home builder discount rates are lower because the properties they purchase are generally ready to build, whereas the subject property requires infrastructure development prior to vertical construction. (Id. at 64.) Balancing the above rates, Call concluded a discount rate of 20 percent was appropriate. (Id.) Next, Call took the anticipated income stream over six years with four percent appreciation, added TDT credits and LID reimbursements from the City, subtracted the allocated infrastructure costs and discount rate to arrive at a value of $28,760,000.

Call testified that he had prepared other appraisals on all or part of the Reed’s Crossing property in August 2014 (Ex G), December 2016 (Ex H), February 2017 (Ex I), and October 2017 (Ex J). He testified that his appraisal for the subject property as of January 1, 2017, is the most accurate opinion of value as of the assessment date. He further testified that the other appraisals were prepared for other purposes—such as internal decision making and were done quicker and with less precision. Additionally, Call testified those appraisals did not define the same area parameters as those in the appraisal submitted by Plaintiffs in Exhibit 1.

Call testified that Defendant’s approach to value—estimating revenue of $85 million minus costs to come up with $67 million—does not capture the timing of the costs relative to development (actual construction), does not account for timing of revenue and expenses, the risks of obtaining entitlements, final approvals, and unknown conditions that may arise during construction. The retail values used by Defendant assume that infrastructure—streets, water, power, drainage—are already in place and that vertical construction could be accomplished. As to TDT credits, Call asserted that a market participant would estimate the amounts as he did in his appraisal. With respect to the projected building density used by Defendants, Call used an

DECISION TC-MD 180141R 10 average density not maximum density because in his experience “zoning does not dictate density, the market does.” B. Defendant’s Evidence of Value Morrison testified he has been the Development Services Manager in Public Works for the City since June 2017. He oversees the administration of development activities for Public Works, which includes permitting, Geographic Information System (GIS) and certain transportation funding tools/fees. Three funding tools that impact the subject property are the Transportation Development Tax (TDT), the Transportation System Development Credits (TSDC), and the Local Improvement District (LID). Washington County offers these credits to be earned by developers when doing certain capacity improvements, agreed by the City, in connection with their development. The credits can be applied to development costs and can be sold or transferred to others up to ten years after issuance, however, unused credits lapse. Morrison testified that certain criteria that must be met to qualify for the credit and to determine the percentage of credit applicable. Morrison acknowledged that although the standards for determining the reimbursement were in place, the conditions were not met as of January 1, 2017.

Vai testified she has worked as an appraiser for Washington County Assessment and Taxation since November 2018 and holds a general commercial license. She formally worked for PGP Valuation which became Colliers and then CBRE. Vai performed a retrospective appraisal of the subject property as of the assessment date. (Ex A.) She noted that although Plaintiff and the City have an overall plan for the subject property, the plan is flexible. Vai testified she made an extraordinary assumption that the subject property was an independent lot, rather than part of a larger tract. She testified that the housing market in the Hillsboro and Portland Metropolitan areas are “very hot” and proximity to high paying jobs makes the subject

DECISION TC-MD 180141R 11 property more valuable.

Vai testified that she looked at the subject parcel by zoning component, because that is what the market would do. Vai testified that market participants would treat the subject property as already partitioned and under contract even though technically, the lots were not ready for development. To counter Plaintiff’s assertion that the lots could not be sold, she observed that Plaintiff had sold a 40-acre parcel to the school district, with location to be determined.

Vai agreed that market participants would likely use the sales comparison approach as modified by the DCF analysis. However, Vai preferred the sales comparison approach alone, because the DCF requires too many assumptions. She believes that the approaches should yield similar results, and because Plaintiff’s values between the sales comparison approach and DCF were dramatically different, then their conclusions are unreliable.

1. Defendant’s sales comparison approach Vai focused on 107 developable acres and reduced that figure in the analysis by ten percent for streets. She testified that one big difference for her appraisal is the importance of zoning density to value. Vai did not believe the SFR-4.5 and SFR-6 zones should be lumped together the way Plaintiff did because in her view developers would try to maximize density to minimize cost per unit and maximize profit. Vai’s report stated that based on the City code, the zoning would allow for 2,007 to 2,484 units on the subject property. Vai reviewed density data for Butternut Creek (phase 1 & 2) and Rosedale Parks (phase 2 & 3) to determine the density market participants were building at and concluded an average density of approximately 90 percent of maximum allowable density. (Ex A at 49.) Vai later clarified that her it should have been calculated at 90 percent of the average density builders in the area are actually building. ///

DECISION TC-MD 180141R 12

a. SFR 4.5

Vai selected six comparable land sales. (Ex A at 64.) Comparable #1 was the sale of 58.5 acres of development land at $649,573 per acre with a density of 8.48 lots per acre. (Id.) Some entitlements were done, but the buyer had to get their own permits. (Id. at 66.) The price was negotiated in August 2014 and the deed was recorded one year later. Comparable #2 was the sale of 29.22 acres near Comparable #1 at $499,658 per acre with a density of 5.92 lots per acre. This lot was purchased without entitlements, in an inferior market, and had terrain and wetland issues. (Id. at 66-67.) Comparable #3 was the sale of 31.4 acres in Portland at $496,105 per acre with a density of 4.62 lots per acre. Vai testified the purchaser decided to wait for better market conditions prior to developing the property. Comparable #4 was the sale of 36.42 acres in Rosedale Parks (Hillsboro) at $496,105 per acre with a density of 7.22 lots per acre. Comparable #5 was the sale of 31.72 acres in West Hills (Beaverton) at $533,346 per acre. (Id. at 64.) The purchaser built a 192-lot development that is near other planned developments and adjacent to Comparable #2. (Id. at 63, 68.) Comparable #6 was the sale of 34.29 acres in North Plains at $460,871 per acre with a density of 8.81 lots per acre. (Id. at 64.) This lot had terrain issues requiring retaining walls.

Based on her concluded average density of 90 percent of maximum (Ex A at 49), and a density range of 8 to 10 lots per acre for SFR-4.5, Vai concludes that the density would be 9.01 for the subject property. (Ex A at 65.) Vai did quantitative analysis instead of Call’s qualitative analysis—adjusting for location, size, shape, topography, density, entitlements, and time. (Id. at 69.) Based on data from the RMLS, property in the area of the subject property was increasing at an average of 11.1 percent over a five-year period. That rate roughly matched Defendant’s records for residential land at 10 percent per year, and thus Vai used the lower figure. Vai also

DECISION TC-MD 180141R 13 adjusted for physical characteristics. (Id. at 64.) She adjusted 25 percent downward for entitlements to Comparable #1. (Id. at 69.) She adjusted for density for example, Comparable #2 that had 5.92 units per acre compared to the subject’s 9.01, which means the lots in Comparable #2 would be bigger and generate less revenue for the developer. To verify her hypothesis, Vai looked at the range homebuilders are paying for each unit at $50,000 to $105,000 and settled on an adjusted price per unit of $64,000. (Ex A at 69-70.) Vai determined that 213 lots could be built yielding a rounded per lot value of $64,959 equating to a price per acre rounded to $649,594. (Id. at 70.) With that value multiplied by the number of acres, the total value of the SFR-4.5 was determined to be $13,850,000. (Id.)

Vai criticized Plaintiff’s Comparable #1 as being in an inferior market, and Comparable #3 as not relevant because it was not yet inside of the Urban Growth Boundary. She felt that Comparable #2 was a reasonable choice but the location and distance from OR-217 did not justify categorizing it superior to the subject property.

b. SFR-6

Vai created a separate analysis for the SFR-6 category because of the density difference.

She selected six sales in her analysis of the SFR-6 zone. Vai used a density of 6.75 units per acre for the SFR-6 land. (Ex A at 72.) She acknowledged that smaller lot sizes increase the price. Vai adjusted for time, size, topography, and density. (Id. at 76.) Comparable #2 was a 5-acre tract sold for a high-density townhouse development. (Id. at 74.) Vai’s net adjustments for that comparable were 47 percent. (Id. at 76.) Comparables #4 and #6 received similar large adjustments of 30 percent and 33 percent respectively. (Id.) Comparable #3 had the least adjustment at 10 percent. (Id.) Vai placed the greatest weight on Comparables #3 and #4 because of their similar in size and design to the subject property. (Id.) She opined that 40 lots

DECISION TC-MD 180141R 14 could be built indicating a range of value of $2,886,840 to $2,967,030 and concluded that the rounded value for the subject property’s 5.94 acres was $496,633 per acre for a total indicated value of $2,950,000. (Id.)

c. MFR

Vai considered the multifamily, MFR-1, MFR-2, and MRF-3 zones together, and used a range of 10 comparable sales. (Ex A at 79.) Vai considered the density range of these zones to be from 11 to 28.75 dwelling units per acre. (Id. at 78.) In the appraisal report, Vai opined that developers would base their projections on 90 percent of allowable density to maximize their profit. (Id.) Based on that methodology, Vai estimate that MFR-1 would have a density of 14.4 units per acre and a total value of $5,200,000. (Id. at 86.) For the MFR-2, Vai estimated that it would have a density of 19.15 units per acre for a total value of $11,500,000. (Id.) Lastly, for the MFR-3, Vai estimated it would have a density of 25.87 units per acre for a total value of $5,600,000. (Id.) Vai concluded that her approach to the MFR zones was superior to Plaintiff’s because she broke them into components based on different permissible densities, whereas Plaintiff’s appraisal lumped them all together. In reviewing Plaintiff’s comparable sales, Vai testified that she would not have used a Happy Valley sale because that market area is very different. (Comparable #3, Ex 1 at 50). She also did not use Plaintiff’s Comparable #5 because in her view it was an “outlier.”

d. MU-VTC

For the MU-VTC, Vai’s research indicated that commercial land would be sold in smaller increments and that apartments would fill in the remainder, restricted only by parking limitations. (Ex A at 88.) The 50.47-acre zone allows for a variety of uses which Via broke up into one-third commercial and two-thirds multi-family housing. (Id. at 87.) As in the other

DECISION TC-MD 180141R 15 zones, Vai assumed a 90 percent density, which would allow for 1,322 net units. (Id.) Vai calculated that the multi-family component of this zone would yield about $1 million per acre for a total of $30,500,000. (Id. at 88.) Vai found six commercial sales with adjusted prices ranging from $908,154 and $1,115,640. (Id. at 94.) Vai concluded that the price per acre was $1,051,180 resulting in an indicated value for this zone of $5,300,000. (Id. at 95.) Adding the two components together yield an aggregate value for the zone at $15.9 million. (Id.)

2. Sales Comparison Approach Totals After concluding on each of the zones separately, Vai added the values for each component zone which equaled $85,650,000. (Id. at 96.) Vai testified she spoke a number of national developers and formed the opinion that a 20 percent discount would be applied if a purchaser bought the entire property as a whole. With the discount, Vai arrived at a concluded market value of $67,300,000. 1 Vai did not take into account “extraordinary costs” because she did not have those figures. Vai did not include LID costs, because she felt the costs and reimbursements “nets out to zero.”

3. DCF Analysis Vai used estimated absorption figures of two to three months for the SFR zone, three to five months for the MFR zone, and six to nine months for the MU-VTC. (Ex A at 98.) Vai considered holding expenses of three years, which she considered minimal, marketing/sales commissions at five percent, and a discount rate for planned unit developments in the pacific northwest region ranging for 14.91 to 33.46 percent and averaging 22.25 percent. (Id. at 99- 100.) She did not consider costs and reimbursements because she was not provided with those figures and she opined that they probably would be “a wash” anyway. She concluded that with a

1 This is different that the value on Ex A at 97 due to a calculation error.

DECISION TC-MD 180141R 16 12-quarter sell-out period that a 24.1 percent discount rate was appropriate, resulting in a bulk value of $67,300,000 million.

On cross-examination, Vai clarified that her density computation on Exhibit A at 49 was that builders were building to 90 percent of average density, not maximum density. (See Tr at 464-65). She testified that, after recalculating with the correct figures, the data supports 77 percent of maximum allowable density instead of 90 percent. (Id. at 468.) She stated that the 20 percent discount rate used above captures risk, but not costs of the developer. (Id. at 473.) Vai admitted she did not adjust for LID because she had “insufficient data.” (Id. at 490.) She testified that master planned communities are subject to higher costs, a longer time frame of development, and are subject to more governmental approvals, than a subdivision. (Id. at 501.)

II. ANALYSIS

This appeal involves a select 140-acre portion of a larger development site known as Reed’s Crossing, which in turn is a subset of a larger planned community development known as South Hillsboro. As of the assessment date, January 1, 2017, the subject property was vacant land, with some grading and a zoning plan in place, but without entitlements that would allow vertical development. Prior to building, a number of conditions needed to be met which both parties agree were not in place as of the assessment date. The parties considered and rejected the cost approach and relied on the sales comparison approach while offering different modifications using a variation of the income approach known as the Discounted Cash Flow (DCF) analysis.

The court is asked to determine the real market value of the subject property. Real market value is defined in ORS 308.205(1), 2 which states:

“Real market value of all property, real and personal, means the amount in cash that could reasonably be expected to be paid by an informed buyer to an informed

2 The court’s references to the Oregon Revised Statutes (ORS) are to 2015.

DECISION TC-MD 180141R 17 seller, each acting without compulsion in an arm’s-length transaction occurring as of the assessment date for the tax year.”

The assessment date for the 2017-18 tax year is January 1, 2017. ORS 308.007; ORS 308.210. The real market value of property “shall be determined by methods and procedures in accordance with rules adopted by the Department of Revenue * * *[.]” ORS 308.205(2). The three approaches to value that must be considered are: (1) the cost approach; (2) the sales comparison approach; and (3) the income approach. OAR 150-308-0240(2)(a). Although all three approaches must be considered, all three approaches may not be applicable in each case. Id. Here, both parties relied primarily on the sales comparison approach and then overlaid an income approach, the DCF, to adjust the value. Using the sales comparison approach, the “court looks for arm’s length sale transactions of property similar in size, quality, age and location” to the subject property. Richardson v. Clackamas County Assessor, TC-MD 020869D, 2003 WL 21263620 at *3 (Or Tax M Div, Mar 26, 2003). Plaintiff bears the burden of proof and must establish its case by a preponderance of the evidence. ORS 305.427. “[T]he court has jurisdiction to determine the real market value or correct valuation on the basis of the evidence before the court, without regard to the values pleaded by the parties.” ORS 305.412. A. Sales Comparison Approach The parties generally agree that the sales comparison approach, divided into several zoning sections is the best starting point for valuing of the subject property. The parties disagree on how finely to categorize and analyze the many differently zoned sections and whether the sales comparison approach standing alone is the best method to determine value. Plaintiff combined the SFR zones into a single analysis and averaged them. They followed the same approached for the MFR and MU-VTC zones. Defendant broke out each zoning type individually and found comparable sales for each category. Defendant’s analysis focused on the

DECISION TC-MD 180141R 18 probable density as the main factor in determining value of those zones.

Both parties acknowledged that there were no comparable sales in the general area matching the size and raw condition of the subject property. Plaintiff approached the challenge by finding comparable sales in zone categories of SFR, MFR, and MU-VTC, and adjusting them qualitatively. Defendant approached the challenge by finding comparable sales for zone categories of SFR-4.5, SFR-6, MFR-1, MFR-2, MFR-3 and MU-VTC and adjusting them by looking at average density of homes being built in the area and their associated sales price, and using that figure to arrive at a price per acre. Defendant also adjusted the sales prices quantitatively. The court views each parties’ approach as viable methods for determining the value of subject property. That said, in this instance the court views Plaintiff’s sales comparison approach, averaging the zones, as superior given the raw undeveloped condition of the subject property as of the assessment date for the reasons set forth below.

1. Combining zoning areas In theory, Defendant’s approach promises greater precision because the density and value of each of the zones are different. However, the parties each independently acknowledged that although zoning areas have been approved, there will be further changes to zoning and density limits, based topography, market demand, and agreements with the City. The court is persuaded by Plaintiff’s witnesses that market participants seeking to purchase the subject property at this early stage of development would account for the uncertainty by grouping zones together.

2. Density calculations based on zoning During the trial and in their closing brief, Defendant acknowledged errors in their appraiser’s calculations. Specifically, Vai reviewed the average density that homebuilders were actually building in nearby areas and found they were building at 90 percent of average density

DECISION TC-MD 180141R 19 or 77 percent of the maximum allowable density. The court agrees that builders in the vicinity of the subject property are currently building at nearly 80 percent of maximum allowable density. But the evidence was not persuasive that potential purchasers of the subject property would negotiate their purchase price based on the higher density figures. Universally, witnesses testified that potential buyers/home builders scrupulously analyze expenses. It is hard, then, to accept that those individuals would make their purchase assumption at, or near, the maximum possible densities because any problems with entitlements, geology, or topography would diminish their profitability. See Appraisal Institute, The Appraisal of Real Estate 348 (15th ed 2020) (“Without surveys and engineering studies, an appraiser cannot know exactly how many lots can be created from a particular parcel of land.”) Defendant’s assumption of near maximum density results in an overly optimistic valuation because at the date of assessment the property was still in the initial stages of development. To put this another way; the stage of development as of the assessment date is too early for the appraisers to be determining value by counting doors. The court is persuaded by Plaintiff’s witnesses that market participants seeking to purchase the subject property at this early stage of development would not base their purchase price on zoning maximization.

3. Adjustments to Plaintiff’s comparable sales Plaintiff’s comparable sales for the SFR zoning areas appear reasonable and the court adopts its value of $425,000 per acre. A majority of Plaintiff’s comparables for the MFR residential development zone appear reasonable. However, Comparable #3 in Happy Valley and Comparable #5 purchased by Newland Communities (Nash Holland) are problematic. The court agrees with Vai’s assessment that the market in Happy Valley is not similar to South Hillsboro and thus it should be given less weight. As to Comparable #5, Defendant’s challenged the nature

DECISION TC-MD 180141R 20 of the sale as not being an arm’s-length transaction. Although Brentlinger and Call testified that they believed this sale represented an arm’s-length transaction because the parties had a motive to maximize their own self-interests, that explanation is not convincing. The court finds Comparable #5 unreliable for three reasons, first, the sale is between related parties. Mr. Holland the purchaser has an interest in Plaintiff. Second, the sale at $373,184 per acre is substantially lower than all the other comparables; more than 15 percent below Call’s MFR value and more than 45 percent below Call’s MU-VTC value. Third, the land is zoned MFR-1, MFR- 3, and MU-VTC which would potentially make it more valuable than other MFR land given its flexible commercial component. Those factors cast doubts as to the arm’s-length nature of the sale and whether it is a good comparable. See OAR 150-308-0240(2)(c) (“In utilizing the sales comparison approach, only actual market transactions of property comparable to the subject, or adjusted to be comparable, may be used.”) Plaintiff’s testimony does not overcome those doubts. Thus, the court rejects that comparable. The court gives greater weight to Comparable #4 which was a Beaverton sale at approximately $550,000 per acre. The court agrees with Call that this sale is a slightly higher indicator of value. The court finds the value of MFR zone at $525,000 per acre.

Call estimated the value of the MU-VTC at a rounded value of $700,000 per acre after considering four comparable sales ranging from $16 square foot to $26 square foot. Call determined that the subject should be valued on the lower end of the range because of high LID costs of $3 per square foot. In reviewing the comparables, the court places little weight on Comparable #1 because it was not marketed and/or purchased for development. The best comparable matching the MU-VTC zoning is Comparable #4, with an adjusted value of $26 square foot. That figure closely matches Call’s October 2017 appraisal which estimates a value

DECISION TC-MD 180141R 21 for the MU-VTC portion at $24 to $26 per square foot. (Ex J at 21-24.) It also matches a contemporaneous letter of intent for the subject property, from a prospective purchaser, mentioned in the October 2017 appraisal. (Id. at 24.) Finally, it closely matches Vai’s value of approximately $24 per square foot ($1,050,000 per acre.) Thus, the court finds that value of the MU-VTC using the sales comparison approach at $1,050,000 per acre. A summary of the values determined by the court using the sales comparison approach are as follows:

Zone Acres Value/acre Total SFR 27.26 $425,000 $11,585,500

MFR 35.03 $525,000 $18,390,750 MU-VTC 45.43 $1,050,000 $47,701,500 $77,677,750

B. The DCF Analysis The parties generally agree that market participants would utilize a DCF analysis to evaluate the real market value of the subject property. Plaintiff asserts that a DCF analysis is essential to account for the cost, time, and risk in completing the infrastructure improvements and the remaining hurtles needed to take place to sell the land, especially with irregular flows of income. Vai asserts that a comparable sales approach is more accurate because the DCF analysis involves too many assumptions. Further she argued that the difference in the values concluded by Plaintiff’s sales comparison approach and the DCF analysis suggest that the DCF analysis results are questionable.

In First Interstate Bank of Oregon, N.A. v. Dept. of Rev., 306 Or 450, 760 P2d 880 (1988), Oregon’s Supreme Court considered whether the “developer’s discount” under a DCF analysis applied to a fully developed subdivision. The court concluded that “[t]he developer’s

DECISION TC-MD 180141R 22 discount does not assess the value of the properties if put to their highest and best use, but reduces their value to arrive at the value of the properties considered as an investment. Investment is not the highest and best use of the properties.” Id. at 455. The court emphasized the fact that the subdivision at issue had “been subdivided and roads and utilities provided to each lot” such that “there [were] no longer development costs that would affect the present market value of each lot.” Id. The court went on to state that:

“Although we reject a developer’s discount in the situation in which the original holding has been subdivided into several lots and the subdivision has been fully developed, it is appropriate to take into account the present legal and physical status of the property. Even if the best use of a property would be for subdivision, the property’s present value would take into account the fact that the property would not be presently usable for the sale of individual lots.”

In Powell St. I, LLC v. Multnomah Cty. Assessor, 365 Or 245, 259-60, 445 P3d 297 (2019), this court considered the holding in First Interstate and concluded that it did not, as argued by the department, “prohibit a valuation methodology that takes into account the property owner’s cost to sell or improve properties” where the properties highest and best use was “as part of a group of lots.” Here, the parties agree that the highest and best use of the subject property was to a developer or group of builders who could invest in the infrastructure and entitlements necessary to begin vertical construction. (Ex 1 at 41, Ex A at 58.) The court agrees it was simply not feasible to sell the subject property in individual lots on the assessment date of January 1, 2017. Accordingly, the court agrees that a DCF analysis is necessary in this case to determine the real market value and that market participants would use this approach. The court attributes the divergent value differences between the sales comparison approach and the DCF analysis to the fact that none of the comparables were easily adjustable to the raw conditions of the subject property. ///

DECISION TC-MD 180141R 23

Defendant’s DCF analysis discounted 20 percent for risk but assumed that 100 percent of Plaintiff’s development costs would be reimbursed and would be a “wash.” Defendant chose a 25 percent discount rate but did not include in the analysis extraordinary costs, because it did not have those figures. The court believes that the market would not ignore such a potentially significant factor such a multi-million-dollar investment. It leaves the analysis incomplete.

As between the parties’ variations, the court generally agrees with Plaintiff’s approach.

Call’s assumption of a four percent appreciation rate appears reasonable as do the assumptions of an initial sale in 2018 and subsequent phases in 2020 and 2022. The court adopts Plaintiff’s estimated a 5-year sellout period with future TDT credits and reimbursements at 50 percent or $9.7 million. Morrison testified that reimbursement is from 50 to 100 percent depending on several factors, and that he would decide after Plaintiff filed an application. At this early stage of development, a potential market participant would have to estimate the credits. Further, they would not likely pay 100 percent of a future credit now, especially if unused credits were not refundable. For the reimbursements, Plaintiff will receive $10.5 million for the rail crossing, to which Call allocates $6.5 million to the subject property. He allocated payments over five years which also appears appropriate.

Having concluded that that the subject property had a total real market value of $77,677,750 under the sales comparison approach, the court applies Plaintiff’s DCF analysis (Attachment 1) 3, resulting in a 2017-18 real market value of $40,398,320. /// ///

3 The court’s DCF calculation corrected what appeared to be errors contained in Plaintiff’s calculations.

While the court initially believed that in theory the SFR and MFR should not be averaged for the calculation, the difference was so proportionally small that the original formula was used.

DECISION TC-MD 180141R 24

III. CONCLUSION

After careful consideration the court finds the real market value of the subject property as of January 1, 2017, was $40,398,320. Now, therefore, IT IS THE DECISION OF THIS COURT that the 2017-18 real market value of the subject property is $40,398,320.

Dated this ____ day of June 2021.

RICHARD DAVIS

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 Richard Davis and entered on June 4, 2021.

DECISION TC-MD 180141R 25

Attachment 1

REVISED DCF ANALYSIS

2017 2018 2019 2020 2021 2022 Residential Land Value (4% appreciation) $ 480,000 $ 499,200 $ 519,168 $ 539,935 $ 561,532 $ 583,993 4% Commercial/Mixed (4% appreciation) $ 1,050,000 $ 1,092,000 $ 1,135,680 $ 1,181,107 $ 1,228,351 $ 1,277,486

Residential Acres Sold 0.00 31.15 31.14 0.00 0.00 0.00 Commercial/Mixed acres sold 0.00 15.15 0.00 15.14 0.00 15.14 Revenue Residential Land $ - $ 15,550,080 $ 16,166,892 $ - $ - $ - Commercial/Mixed $ - $ 16,543,800 $ - $ 17,881,963 $ - $ 19,341,131 TDT Credits $ - $ 1,954,009 $ 1,954,009 $ 1,954,009 $ 1,954,009 $ 1,954,009 LID Reiumbursement Railroad Crossing $ - $ - $ - $ - $ - $ 6,489,570

Allocated Gateway Costs $ (6,796,612) $ (11,124,159) $ - $ - $ - $ - Annual Cashflow $ (6,796,612) $ 22,923,730 $ 18,120,901 $ 19,835,972 $ 1,954,009 $ 27,784,710 20% Net Present Value @ 20 percent $40,398,320.20