Digital 60 & 80 Merritt, LLC v. Board of Assessment Appeals

211 Conn. App. 559

Opinion

*********************************************** The “officially released” date that appears near the beginning of each opinion is the date the opinion will be published in the Connecticut Law Journal or the date it was released as a slip opinion. The operative date for the beginning of all time periods for filing postopinion motions and petitions for certification is the “officially released” date appearing in the opinion.

All opinions are subject to modification and technical correction prior to official publication in the Connecticut Reports and Connecticut Appellate Reports. In the event of discrepancies between the advance release version of an opinion and the latest version appearing in the Connecticut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports, the latest version is to be considered authoritative.

The syllabus and procedural history accompanying the opinion as it appears in the Connecticut Law Journal and bound volumes of official reports are copyrighted by the Secretary of the State, State of Connecticut, and may not be reproduced and distributed without the express written permission of the Commission on Official Legal Publications , Judicial Branch, State of Connecticut. ***********************************************

DIGITAL 60 & 80 MERRITT, LLC v. BOARD OF ASSESSMENT APPEALS OF THE TOWN OF TRUMBULL ET AL.

(AC 44296) Alvord, Prescott and DiPentima, Js.

Syllabus

The defendant town of Trumbull and its Board of Assessment Appeals appealed to this court from the judgment of the trial court sustaining the plaintiff’s appeals from the decisions of the board, which upheld the town’s tax assessments levied against the plaintiff’s real property. In 2010, the plaintiff purchased the real property with the intent of leasing it out as a data center, a location to house and secure electronic data. It organized the building into colocation suites, each of which were occupied by multiple users. The plaintiff provided the space, the raised floors, power, cooling, Internet connectivity, security, and the redundancy required to store the electronic data, and the colocation customers either provided their own computers or leased them from the plaintiff. In 2011, the plaintiff decided to remediate and expand the property to, inter alia, build two additional data suites, Suite 210 and Suite 220. The plaintiff intended each suite to be occupied by a single wholesale customer, who would supply its own computers and racks. The plaintiff, however, was unable to find wholesale customers for the new suites and, by mid-2013, it was leasing space within Suite 210 to colocation customers. The construction on Suite 220 was never completed . By the time of trial, it remained raw space with only an unfinished concrete floor, walls, and a ceiling in place. As a result, the plaintiff claimed that the suite was unfit to be leased even as powered base building (PBB) space, which would require the plaintiff to supply a space with completed exterior construction, power, and connectivity, while the customer would build out the interior to its own specifications. The town assessed the property as part of its revaluation for its 2011 grand list. It then conducted interim reassessments of the property in 2013 and 2014, pursuant to the applicable statute (§ 12-53a), to take into account the new construction. Following these reassessments, the town assessor determined that the fair market value of the property, based on its physical condition as of October 1, 2013, and October 1, 2014, respectively, and market conditions as of October 1, 2011, was approximately $145,446,000. The plaintiff appealed the assessor’s 2013 and 2014 valuations to the board, which denied its appeals. The plaintiff then appealed to the trial court, which found that the fair market value of the property, based on its physical condition as of October 1, 2013, and October 1, 2014, and market conditions as of October 1, 2011, was $109,000,000, and, accordingly, it sustained the plaintiff’s appeals with respect to its claims of excessive valuation. On the defendants’ joint appeal to this court, held:

1. The trial court’s determination that Suite 220 had no income and no income potential in 2011 was not clearly erroneous: a. Contrary to the defendants’ claim, there was evidence in the record to support the trial court’s factual finding that there was no market for Suite 220 in Trumbull in 2011, namely, the testimony of L, the appraiser serving as the plaintiff’s trial expert, and D, one of the plaintiff’s executives , which the court found to be credible. b. The defendant’s argument that Suite 220 clearly added value to the property, as allegedly confirmed by L’s cost approach analysis, rested on a faulty premise: the trial court found that the income capitalization approach, rather than the cost approach, was the most reliable and appropriate valuation method for the property, and the defendants conceded that, under such an approach, the income producing potential of the suite was determinative of its value; accordingly, because there was evidence in the record that there was no actual or market rent for Suite 220, this court could not conclude that the trial court’s finding that no income potential existed was erroneous. c. Despite the defendants’ request, this court declined to usurp the role

of the trial court and reweigh the evidence relating to the market for Suite 220 in Trumbull in 2011 in their favor, as such evidence was before the trial court and carefully considered by it, and it was the role of the trial court to determine the credibility of such evidence. d. Contrary to the defendants’ claim, it was not improper for the trial court to consider evidence of the Trumbull data center market in 2013 or 2014 in determining that there was no market for Suite 220 in Trumbull in 2011: that evidence was relevant because S, the appraiser serving as the defendants’ trial expert, considered information regarding market rent and market conditions through October 1, 2014, in his analysis; moreover, although the trial court referenced the plaintiff’s argument concerning the lack of success in renting the suite within the years immediately prior to trial, it made clear that such argument was not material to its decision and that it instead relied on evidence of the data market in Trumbull in 2011, 2013 and 2014 in determining that there was no market for Suite 220 as of the revaluation date; furthermore, the defendants failed to provide this court with any legal support for their argument that the trial court’s consideration of the market through S’s 2014 cutoff date was improper; accordingly, the trial court did not err in concluding that none of the evidence credibly supported the defendants’ claim that Suite 220 was marketable as PBB space in 2011, 2013 and 2014. e. Contrary to the defendants’ assertion, the trial court’s finding that there was no market for Suite 220 was not inconsistent with its market rent analysis: evidence in the record, namely, the plaintiff’s investment committee memoranda and D’s testimony, supported the trial court’s conclusion that any market for the property would lie exclusively with Connecticut based customers; moreover, although the court found that the Trumbull market was comparable to the Boston market with respect to the size of potential tenants, it determined that demand in Trumbull was limited by the inherent characteristics of the Connecticut market, which finding was not unfounded or contrary to the trial court’s market rent analysis. f. Despite the defendants’ claim, there was evidence in the record indicating that Suite 220’s square footage made it unmarketable as PBB space, including D’s and L’s testimony, which the trial court found to be credible, and the plaintiff’s form 10-K annual reports, filed with the Securities and Exchange Commission in 2011 and 2013, which included information regarding the average number of total square footage of the plaintiff’s new and renewal PBB leases.

2. There was sufficient evidence in the record to support the trial court’s determination that the highest and best use of Suite 210 was as wholesale space: although the trial court credited S’s testimony that colocation was becoming more popular, it determined that such testimony did not preclude it from agreeing with L’s conclusion that the highest and best use of the suite was as wholesale space, especially in light of D’s testimony regarding the various factors that made colocation rentals less desirable than wholesale rentals; moreover, contrary to the defendants’ assertion, the trial court properly considered Suite 210’s actual rent in determining its fair market value, however, because the existing leases were for colocation customers, the court applied market rental rates to determine the suite’s fair market value as wholesale space; furthermore, the court’s conclusion that the highest and best use of Suite 210 was as wholesale space did not contradict its conclusion that there was no market for PBB space in Trumbull in 2011, as such conclusion did not preclude the court from also finding that there was demand for a wholesale lease of a fully built out suite, like Suite 210.

3. The trial court’s application of a capitalization rate of 8 percent to the property’s net income in determining its market value was supported by evidence in the record: the trial court determined that the capitalization rate proposed by L was essentially correct; moreover, in making its determination, the trial court properly gave credence to one appraiser’s method of calculating the capitalization rate over the other’s and determined an appropriate capitalization rate based on the evidence in the record, which included information regarding capitalization rates for various other transactions throughout the country that were detailed in industry newsletters and reports; accordingly, to arrive at its capitalization rate, the trial court adjusted the rate proposed by L to ensure that sufficient consideration was given to the effect of the electrical infrastructure installed on the property during its remediation and expansion .

4. Contrary to the defendants’ assertions, the trial court’s alleged failure to consider the plaintiff’s internal property valuations did not make its determination of the fair market value of the property clearly erroneous: the trial court determined that the plaintiff’s 10-K annual reports for 2013 and 2014 reflected the book value of the property, rather than the fair market value, and, as a result, such statements were not relevant to its calculations; moreover, the trial court determined that the valuation set forth in the plaintiff’s 2014 impairment analysis of the property, which was based on optimistic assumptions and a best case scenario and included the value of various items of personal property located on the property, was consistent with its own valuation; accordingly, the trial court considered, but did not place weight on, the plaintiff’s internal valuations.

Argued December 1, 2021—officially released April 5, 2022

Procedural History

Appeals from the decisions of the named defendant upholding the valuations made by the assessor of the defendant town of Trumbull of certain of the plaintiff’s real property, brought to the Superior Court in the judicial district of Fairfield and transferred to the judicial district of New Britain, where the appeals were consolidated and tried to the court, Hon. Stephen F. Frazzini, judge trial referee; judgments sustaining the appeals, from which the defendants filed a joint appeal with this court. Affirmed.

Mario F. Coppola, with whom were Matthew L. Studer and, on the brief, Gregory S. Kimmel, for the defendants (appellants).

Charles D. Ray, with whom were Shawn S. Smith and, on the brief, Angela M. Healey, for the plaintiff (appellee).

Opinion

DiPENTIMA, J. In this joint real estate tax appeal, the defendants, the Board of Assessment Appeals of the Town of Trumbull (board) and the town of Trumbull (town), appeal from the judgments of the trial court sustaining the appeals1 brought by the plaintiff, Digital 60 & 80 Merritt, LLC,2 and ordering the reduction of the defendants’ tax assessment levied against the plaintiff’s property located at 60 Merritt Boulevard in Trumbull (property). On appeal, the defendants challenge the court’s determination of the fair market value of the property, which they claim is based on certain clearly erroneous factual findings made by the court. Specifically , the defendants claim that the court erred in (1) failing to impute income to Suite 220 of the property, (2) valuing Suite 210 of the property at the suite’s wholesale rate, (3) applying a capitalization rate of 8 percent, and (4) disregarding the plaintiff’s internal valuations. We affirm the judgments of the trial court.

We begin by setting forth the following relevant facts, as found by the trial court, in addition to the relevant procedural history.3 The property includes five acres of land on which a building of approximately 200,000 square feet4 sits (building). The building is connected to a building on another property, located at 80 Merritt Boulevard in Trumbull, which is a separate tax parcel not subject to this appeal. The buildings share a common entrance, security station, and parking lot.

The property was originally the site of a printing company facility, developed in the late 1960s. NASDAQ Stock Market, Inc. (NASDAQ), bought the property in 1996, demolished the printing facility, and, in 1997, constructed the building, a corporate data center, on the site. NASDAQ connected the building to the building located at 80 Merritt Boulevard, which it also owned at the time. In August, 2006, NASDAQ sold both the property and 80 Merritt Boulevard to Sentinel Properties -Trumbull, LLC (Sentinel). Sentinel then began converting the property into a multi-tenant data center facility . In 2010, the plaintiff bought both the property and 80 Merritt Boulevard. The plaintiff undertook an expansion of the property, constructing a 70,000 square foot addition to the building, which was completed in late 2012 or early 2013.

A ‘‘data center’’ is a building, or a portion thereof, that various organizations use to house and secure electronic data. Data is stored on computers, often stacked several shelves high in storage racks. These racks sit on a ‘‘ ‘raised floor’ ’’ that is constructed of perforated tile, sitting approximately three feet above the actual floor. Between the raised floor and the actual floor, power and cooling is installed to supply the computers. Cooling is a vital feature of a data center, as it prevents overheating, which can be caused by the large amount of power supplied to the computers. Another imperative characteristic of a data center is ‘‘ ‘redundancy.’ ’’ Redundancy refers to the inclusion of a second source of power, or backup power, to run the computers and cooling systems in the event that the primary source of power fails. In other words, components of the electrical system are duplicated to prevent the computers from going off-line.

A data center can be designed with individual rooms referred to as data suites. The data center located at the property is organized into individual data center suites. An individual data suite within a data center may have different occupancy arrangements, such as ‘‘wholesale’’ or ‘‘colocation.’’ A wholesale suite is occupied by a single user that provides its own computers and racks, while the building owner supplies the space, raised floor, power, cooling, internet connectivity, security , and redundancy. A colocation suite, on the other hand, is occupied by multiple users. Colocation customers may provide their own computers and racks, or lease them from the building owner. Like wholesale customers, colocation customers rely on the building owner to supply the space, raised floor, power, cooling, internet connectivity, security, and redundancy.

Space within a data center can also be leased as ‘‘powered shell space,’’ which is a data suite that is not fully ‘‘built out.’’ Rather, the space is a data suite ‘‘ ‘with exterior construction completed, available power and connectivity, but with the interior left as raw space to be finished by the customer.’ ’’ The plaintiff has trademarked this type of space as ‘‘powered base building’’ (PBB) space. When a customer leases PBB space, it builds out the space to its own specifications while the building owner provides internet connectivity, power, security, and redundancy.

Sometime before 2010, the plaintiff privately approached Sentinel about purchasing its New England operations, despite the fact that none of Sentinel’s properties was for sale on the open market at the time. In 2010, the plaintiff and Sentinel entered into a contract whereby the plaintiff agreed to pay Sentinel $375 million for two data center properties in Massachusetts in addition to both the property and 80 Merritt Boulevard in Trumbull. The sales price included the plaintiff’s purchase of the leases at each property, the workforce in place, rights of expansion, a time limited noncompete agreement with Sentinel, and goodwill.

Although the plaintiff believed at the time of purchase that the property had full redundancy—the capacity to continue to provide power and cooling if there was a power outage impairing the primary source of power— an off-site explosion in early 2010 revealed that this belief was misguided. The redundancy system did not perform as anticipated, leaving the property without power for a short time. The plaintiff hired a third party to evaluate its electrical system, at which point the plaintiff learned that the system was undersized and was incapable of providing the amount of power that the plaintiff was contractually obligated to provide to existing customers and that was necessary to operate the facility safely. Following the third-party evaluation, the plaintiff learned that the system did not offer any redundancy. This posed a serious problem for the plaintiff . The lack of redundancy violated lease commitments and subjected the plaintiff to rent credits, reputational harm, and the possibility of lost customers.5 The plaintiff decided to replace the existing electrical system with its own system, referred to as a Turnkey Data system. The plaintiff originally estimated that the project would cost approximately $20,250,000. The budget was subsequently increased to $33,521,442 to account for higher than expected engineering and construction costs and for costs paid to the utility company to complete off-site work.

In 2011, the plaintiff also decided to build a 70,000 square foot expansion to the property that would contain three data suites—one on the first floor (Suite 110) and two on the second floor (Suites 210 and 220). The original budget for the expansion was $27,687,500. An internal memorandum describing the expansion stated that the new space would ‘‘ ‘be positioned to attract financial tenants from Fairfield County’ ’’ and noted that a current tenant in the original building was considering moving into the expansion space once it was built and occupying two suites therein. The memorandum also provided that the plaintiff projected ‘‘ ‘an average return on incremental cost of approximately 20 percent’ ’’ from the expansion, which would ‘‘ ‘[help] increase the overall building returns which [would] be diluted by 18 percent because of the [electrical] remediation project. By building the expansion, [the plaintiff] [would] strive to increase the building [return on investment] over time to what it is today, roughly 10 percent.’ ’’ The expansion budget was also later increased to $52,962,157. The electrical remediation and the building expansion were completed by late 2012 or early 2013.

The plaintiff projected that one data suite in the expansion space would be occupied in the first year after construction was completed and would produce additional annual net operating income of more than $2 million annually. The plaintiff also projected that two other data suites in the expansion would be occupied by the end of the second year after construction was completed, resulting in additional annual net operating income of $6 million beyond that received from the original building. As the trial court noted, ‘‘[t]hese rosy projections, however, proved unfounded.’’ One tenant did move from the original building to the first suite in the expansion space, Suite 110, in 2013 but did not occupy two suites as the plaintiff had hoped. Additionally , the plaintiff was unable to find a wholesale tenant to lease the second built out suite in the expansion space, Suite 210. By mid-2013, the plaintiff began leasing Suite 210 to colocation tenants. The third data suite within the expansion space was never completely built out, having only unfinished concrete floors, walls, and a ceiling in place.

The court explained: ‘‘These disappointed expectations , including the need to rent the second expansion suite as colocation space to gain tenants, led [the plaintiff ] to include [the] property in a 2014 review of ‘potentially underperforming’ properties in its portfolio.’’ As part of this internal review, the plaintiff completed a ‘‘ ‘dynamic valuation’ ’’ of both the property and 80 Merritt Boulevard by completing a ten year cash flow analysis based on certain optimistic assumptions. The review resulted in a valuation of $120 million for the two properties . The plaintiff’s dynamic valuation process then required the property to be classified into one of three categories—hold, further analysis, or sell. The two properties were placed into the hold category.

The town typically conducts town wide revaluations of real property in the town every five years, as such revaluations are required by state law. See General Statutes § 12-62 (b) (1). A revaluation establishes ‘‘the present true and actual value’’ of property as of a specific assessment date. General Statutes § 12-62 (a) (5). Towns use that information for the purpose of levying property taxes for the tax assessment year of the revaluation and also for each subsequent tax assessment year that follows, until the next revaluation. General Statutes § 12-62 (b) (1). The town conducted a town wide revaluation effective for the October 1, 2011 grand list.6 The 2011 revaluation assessment, therefore, was applicable to the 2013 and 2014 tax years, the years at issue in this tax appeal.7 Pursuant to General Statutes § 12-53a, however, the town was permitted to conduct an interim reassessment and change its 2011 assessment to take into account the new construction on the property—the electrical remediation and the construction of the expansion— which occurred after the revaluation date. See General Statutes § 12-53a (a) (1) (‘‘[c]ompleted new construction of real estate completed after any assessment date shall be liable for the payment of municipal taxes based on the assessed value of such completed new construction from the date the certificate of occupancy is issued or the date on which such new construction is first used for the purpose for which [the] same was constructed, whichever is the earlier, prorated for the assessment year in which the new construction is completed’’); see also ZML 301 Tresser Ltd. Partnership v. Stamford, 67 Conn. App. 697, 699–700, 789 A.2d 538 (‘‘§ 12-53a . . . authorizes an interim reassessment if a taxpayer has made physical improvements to the property’’ (footnote omitted)), cert. denied, 260 Conn. 902, 793 A.2d 1091 (2002).

For the 2013 and 2014 tax years, the assessor of the town (assessor) was thus tasked with determining the fair market value8 of the property based on physical conditions of the property as of October 1, 2013, and October 1, 2014, but based on market conditions as of October 1, 2011. The assessor concluded that the fair market value of the property as of October 1, 2013, and October 1, 2014, was approximately $145,446,000.9 The assessor also concluded that the fair market value of 80 Merritt Boulevard was $18,038,429 as of October 1, 2013, and $18,137,429 as of October 1, 2014. The plaintiff appealed all four of these valuations and assessments to the board. The appeals were denied.

The plaintiff then filed two separate appeals from the decisions of the board in the Superior Court; see footnote 1 of this opinion; contesting the valuations of both the property and 80 Merritt Boulevard on the town’s 2013 and 2014 grand lists and the board’s refusal to reduce the corresponding tax assessments. In the first appeal, the plaintiff contested the valuations of the property and 80 Merritt Boulevard on the town’s 2013 grand list and, in the second appeal, the plaintiff contested the valuations of the property and 80 Merritt Boulevard on the town’s 2014 grand list. The plaintiff’s complaints sounded in six counts: (1) excessive valuation of the property, (2) disproportionate tax burden of the property, (3) wrongful assessment of the property , (4) excessive valuation of 80 Merritt Boulevard, (5) disproportionate tax burden of 80 Merritt Boulevard, and (6) wrongful assessment of 80 Merritt Boulevard. The parties subsequently stipulated that, for the purposes of the assessment of 80 Merritt Boulevard on the town’s 2013 and 2014 grand lists, that tax parcel had a fair market value of $10,650,000. The plaintiff accordingly withdrew counts four, five, and six of its tax appeals, which related to the 80 Merritt Boulevard property .

Prior to trial, both the plaintiff and the defendants hired tax appraisers to complete appraisal reports of the property. The plaintiff hired John Leary of Advisra Consulting, LLC. The defendants hired Russell Sterling of Sterling, DiSanto, and Associates. These appraisals were ‘‘retrospective’’ appraisals, meaning the appraisals determined the property’s value as of a date in the past. Appraisal Institute, The Appraisal of Real Estate (12th Ed. 2001) p. 54. The appraisers sought to determine the property’s fair market value based on physical conditions of the property as of October 1, 2013, and October 1, 2014, but market conditions as of October 1, 2011. In completing their reports, the appraisers used different cutoff dates.10 The court explained the reason for the differing cutoff dates: ‘‘Sterling maintained that there was ‘no significant difference’ between market conditions for the property on the revaluation date than on

October 1, 2013, and 2014. . . . He thus used information about the building’s tenants, income, and expenses for 2013 and 2014 and also considered information about market rents and conditions through October 1, 2014. Leary testified, on the other hand, that market conditions after 2012 were too different to be considered for an appraisal based on market conditions existing in 2011.’’ (Citation omitted; footnote omitted.) Leary thus established a cutoff date of December 31, 2012.

The appeals from the town’s valuation and assessment of the property on the 2013 and 2014 grand lists were tried before the court over the course of twentythree days between October, 2018, and March, 2019. The issues presented at trial were whether the property had been overvalued and, if so, what was the fair market value of the fee simple interest11 in the property, based on physical conditions of the property as of October 1, 2013, and October 1, 2014, and based on market conditions as of October 1, 2011. At trial, the court heard testimony from Mark DeVestern, the town’s assessor; Sterling, an appraiser serving as the town’s expert; Leary, an appraiser serving as the plaintiff’s expert; David Lucey, the plaintiff’s Vice President Portfolio Manager for the East Region; and William Eaton, a Senior Data Center Manager for the plaintiff. On December 19, 2018, the court conducted a site visit to the property.

In a memorandum of decision dated August 4, 2020, the court concluded that the ‘‘plaintiff [had] proven that the property was overvalued for tax assessment purposes and [was] therefore aggrieved. The court further [found] that the fair market value of the fee simple interest in [the property], reflecting the physical condition of that property as of October 1, 2013, and October 1, 2014, and based on market conditions as of October 1, 2011, was $109,000,000.’’ In accordance with this conclusion , the court sustained the first count of each of the plaintiff’s appeals.12 The court deemed both the second and third counts abandoned for inadequate briefing . The defendants then filed a motion to reargue, reopen, and set aside the judgments of the trial court. The court denied the motion and this joint appeal followed . Additional facts will be set forth as necessary.

We begin by setting forth the well settled legal principles underlying a tax appeal brought pursuant to General Statutes § 12-117a, as well as our standard of review. ‘‘Section 12-117a, which allows taxpayers to appeal the decisions of municipal boards of tax review to the Superior Court, provide[s] a method by which an owner of property may directly call in question the valuation placed by assessors upon his property . . . .’’ (Internal quotation marks omitted.) Ress v. Suffield, 80 Conn. App. 630, 631–32, 836 A.2d 475 (2003), cert. denied, 267 Conn. 920, 841 A.2d 1191 (2004). ‘‘In § 12-

117a tax appeals, the trial court tries the matter de novo and the ultimate question is the ascertainment of the true and actual value of the [taxpayer’s] property. . . . At the de novo proceeding, the taxpayer bears the burden of establishing that the assessor has overassessed its property. . . . Once the taxpayer has demonstrated aggrievement by proving that its property was overassessed , the trial court [will] then undertake a further inquiry to determine the amount of the reassessment that would be just. . . . The trier of fact must arrive at [its] own conclusions as to the value of [the taxpayer’s property] by weighing the opinion of the appraisers, the claims of the parties in light of all the circumstances in evidence bearing on value, and his own general knowledge of the elements going to establish value . . . .’’ (Internal quotation marks omitted.) Aetna Life Ins. Co. v. Middletown, 77 Conn. App. 21, 26, 822 A.2d 330, cert. denied, 265 Conn. 901, 829 A.2d 419 (2003), and cert. denied, 265 Conn. 901, 829 A.2d 419 (2003).

‘‘The goal of property valuation is to determine the present true and actual value of the subject property. . . . The process of valuation at best is a matter of approximation.’’ (Citations omitted; internal quotation marks omitted.) First Bethel Associates v. Bethel, 231 Conn. 731, 738, 651 A.2d 1279 (1995). General Statutes § 12-63 provides in relevant part that ‘‘[t]he present true and actual value . . . shall be deemed by all assessors and boards of assessment appeals to be the fair market value thereof . . . .’’

General Statutes § 12-63b governs the valuation of rental income real property and provides guidance in producing a valuation that best approximates the true and actual value of a property. Section 12-63b (a) provides in relevant part: ‘‘The assessor or board of assessors in any town, at any time, when determining the present true and actual value of real property as provided in section 12-63, which property is used primarily for the purpose of producing rental income . . . shall determine such value on the basis of an appraisal . . . .’’ Section 12-63b (a) specifies three different methods of valuation to determine the true and actual value of real property: (1) the comparable sales approach, (2) the income capitalization approach, and (3) the cost approach. See Walgreen Eastern Co. v. West Hartford, 329 Conn. 484, 497, 187 A.3d 388 (2018); see also Sun Valley Camping Cooperative, Inc. v. Stafford, 94 Conn. App. 696, 702, 894 A.2d 349 (2006).

‘‘A property’s highest and best use is commonly accepted by real estate appraisers as the starting point for the analysis of its true and actual value. . . . [U]nder the general rule of property valuation, fair [market ] value, of necessity, regardless of the method of valuation, takes into account the highest and best value of the land. . . . A property’s highest and best use is commonly defined as the use that will most likely pro-

duce the highest market value, greatest financial return, or the most profit from the use of a particular piece of real estate. . . . The highest and best use determination is inextricably intertwined with the marketplace because fair market value is defined as the price that a willing buyer would pay a willing seller based on the highest and best possible use of the land assuming, of course, that a market exists for such optimum use. . . . The highest and best use conclusion necessarily affects the rest of the valuation process because, as the major factor in determining the scope of the market for the property, it dictates which methods of valuation are applicable. Finally, a trier’s determination of a property ’s highest and best use is a question of fact that we will not disturb unless it is clearly erroneous.’’ (Citations omitted; emphasis omitted; footnote omitted; internal quotation marks omitted.) United Technologies Corp. v. East Windsor, 262 Conn. 11, 25–26, 807 A.2d 955 (2002). Here, both appraisers agreed that the highest and best use of the property was as a data center.

Both appraisers likewise agreed that, in the present case, the only appropriate appraisal methods were the cost approach and the income capitalization approach. ‘‘[N]o one method of valuation is controlling and . . . the [court] may select the one most appropriate in the case before [it].’’ (Internal quotation marks omitted.) Sakon v. Glastonbury, 111 Conn. App. 242, 248–49, 958 A.2d 801 (2008), cert. denied, 290 Conn. 916, 965 A.2d 554 (2009). The court determined, and the defendants do not challenge on appeal, that the proper method for assessing the property in this specific case was the income capitalization approach.

‘‘In the income capitalization approach, an appraiser analyzes a property’s capacity to generate future benefits and capitalizes the income into an indication of present value.’’ Appraisal Institute, supra, p. 471. Under this approach, appraisers ‘‘develop an indication of market value by applying a rate or factor to the anticipated net income from a property. . . . Appraisers arrive at the anticipated net income by considering the property ’s actual rental income, as well as the rental income for comparable properties in the vicinity, property expenses, and allowances for vacancy and collection losses.’’ (Citation omitted; internal quotation marks omitted.) United Technologies Corp. v. East Windsor, supra, 262 Conn. 17 n.9. The income capitalization approach produces a valuation using ‘‘capitalization of net income based on market rent for similar property . . . .’’ General Statutes § 12-63b (a) (2).

‘‘Section 12-63b (b) explains the meaning of market rent as it is used in the income capitalization approach. Specifically, § 12-63b (b) provides: For purposes of subdivision (2) of subsection (a) of this section and, generally , in its use as a factor in any appraisal with respect to real property used primarily for the purpose of pro-

ducing rental income, the term market rent means the rental income that such property would most probably command on the open market as indicated by present rentals being paid for comparable space. In determining market rent the assessor shall consider the actual rental income applicable with respect to such real property under the terms of an existing contract of lease at the time of such determination.’’ (Emphasis omitted; internal quotation marks omitted.) Walgreen Eastern Co. v. West Hartford, supra, 329 Conn. 497. Section 12-63b (b) ‘‘requires that, in determining a property’s market rent, the assessor and, therefore, the court, in determining the fair market value of the property, must consider both (1) net rent for comparable properties, and (2) the net rent derived from any existing leases on the property.’’ (Emphasis omitted; internal quotation marks omitted.) First Bethel Associates v. Bethel, supra, 231 Conn. 740. This approach ‘‘is based on the principle that the amount of net income a property can produce is related to its market value.’’ (Internal quotation marks omitted.) Fairfield Merrittview Ltd. Partnership v. Norwalk, 172 Conn. App. 160, 165 n.9, 159 A.3d 684, cert. denied, 326 Conn. 901, 162 A.3d 724 (2017).

‘‘The income capitalization approach consists of the following seven steps: (1) estimate gross income; (2) estimate vacancy and collection loss; (3) calculate effective gross income (i.e., deduct vacancy and collection loss from estimated gross income); (4) estimate fixed and operating expenses and reserves for replacement of short-lived items; (5) estimate net income (i.e., deduct expenses from effective gross income); (6) select an applicable capitalization rate; and (7) apply the capitalization rate to net income to arrive at an indication of the market value of the property being appraised.’’ Sun Valley Camping Cooperative, Inc. v. Stafford, supra, 94 Conn. App. 702 n.9.

Having set forth the underlying facts and governing legal principles, we turn to our standard of review on appeal. ‘‘We review a court’s determination in a tax appeal pursuant to the clearly erroneous standard of review. Under this deferential standard, [w]e do not examine the record to determine whether the trier of fact could have reached a conclusion other than the one reached. Rather, we focus on the conclusion of the trial court, as well as the method by which it arrived at that conclusion, to determine whether it is legally correct and factually supported. . . . A finding of fact is clearly erroneous when there is no evidence in the record to support it . . . or when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.’’ (Internal quotation marks omitted.) Grolier, Inc. v. Danbury, 82 Conn. App. 77, 78, 842 A.2d 621 (2004). ‘‘In making this determination , every reasonable presumption must be given in favor of the trial court’s ruling.’’ (Internal quotation marks omitted.) Fairfield Merrittview Ltd. Partnership v. Norwalk, supra, 172 Conn. App. 170.

‘‘We afford wide discretion to the court’s determination of the value of property in a property tax appeal. . . . When the court acts as the fact finder, it may accept or reject evidence regarding valuation as it deems appropriate.’’ (Citation omitted.) Grolier, Inc. v. Danbury, supra, 82 Conn. App. 80. ‘‘Because a tax appeal is heard de novo, a trial court judge is privileged to adopt whatever testimony he reasonably believes to be credible.’’ (Internal quotation marks omitted.) Id., 79. Thus, ‘‘credibility determinations are within the exclusive province of the court.’’ Ress v. Suffield, supra, 80 Conn. App. 633.

‘‘Conversely, we review de novo a trial court’s decision of law. [W]hen a tax appeal . . . raises a claim that challenges the propriety of a particular appraisal method in light of a generally applicable rule of law, our review of the trial court’s determination whether to apply the rule is plenary. . . . To be sure, if the trial court rejects a method of appraisal because it determined that the appraiser’s calculations were incorrect or based on a flawed formula in that case, or because it determined that an appraisal method was inappropriate for the particular piece of property, that decision is reviewed under the abuse of discretion standard. . . . Only when the trial court rejects a method of appraisal as a matter of law will we exercise plenary review.’’13 (Internal quotation marks omitted.) Walgreen Eastern Co. v. West Hartford, supra, 329 Conn. 493–94. Keeping these well established principles in mind, we turn to our resolution of the defendants’ claims.

I

The defendants first claim that the court erred in failing to impute income to Suite 220 of the property. Specifically, the defendants claim that the court’s factual finding that Suite 220 had no income potential is clearly erroneous because (1) Suite 220 adds value to the property, (2) the record is devoid of testimony that there was no market for Suite 220 in Trumbull in 2011, (3) Leary’s assertion that there was no market for Suite 220 was not credible and was contradicted by the fact that the plaintiff invested millions of dollars into the suite and actually marketed the suite for lease, (4) the court considered post hoc information, (5) the court’s finding that there was no market for Suite 220 is inconsistent with its market rent analysis, and (6) the record is devoid of any evidence indicating that the suite’s square footage made it unmarketable.14 The plaintiff contends that the court’s decision not to impute any income to Suite 220 is not clearly erroneous because there was evidence in the record that there was no market for Suite 220 in Trumbull in 2011—namely, Leary’s testimony, which the court found credible. We agree with the plaintiff.

The following additional facts are relevant to the resolution of this claim. Suite 220 is located in the expansion space of the property. As the suite existed in 2013 and 2014, it had no raised floor and no sidewalls. At trial, the parties’ experts disagreed as to the nature of Suite 220. Leary opined that the suite was not PBB space because ‘‘ ‘the raised floor and at least the sidewalls would be in place in a [PBB] scenario and not the unfinished state of Suite 220 . . . as it existed in 2013 and 2014 . . . . Money would have to be spent to bring the power and fiber connectivity, the raised floor and the sidewalls to that space in order to lease it as PBB space.’ ’’ Sterling, the defendants’ expert appraiser, categorized the suite as PBB space. The court ultimately concluded that ‘‘nothing in the evidence rule[d] out Sterling’s characterization of the unbuilt suite as PBB space.’’

There was also differing testimony at trial about whether Suite 220 was marketable. As previously noted, both appraisers determined that the highest and best use of the property was its continued use as a data center. Because the highest and best use of the property was determined to be a data center, Leary testified that using Suite 220 for general storage would be impractical considering the security concerns associated with the operation of a data center.15 At trial, Leary testified that he did not attribute any potential income to Suite 220 for the 2013 and 2014 tax years because there was no demand for the suite in Trumbull in 2011. Sterling, however, imputed potential income in the amount of $838,450 per year to Suite 220 for the 2013 and 2014 tax years.

In its memorandum of decision, the court ultimately credited Leary’s assertion that there was no market for Suite 220 in Trumbull in 2011 and concluded that ‘‘[t]he evidence showed that there was no income or potential income from [Suite 220] in 2011 but rather an anticipated 100 percent vacancy rate.’’ We address each of the defendants’ claims attacking this factual finding in turn.

A

The defendants contend that there is no evidence in the record to support the court’s factual finding that, under the market conditions of 2011, Suite 220 had no income potential. We reject this contention.

At trial, during the following exchange between the defendants’ counsel and Leary, Leary explained that he did not allocate any income to Suite 220 based on market rental rates in his appraisal report: ‘‘Q. —with regard to Suite 220, is it correct that you did not attribute any income in your appraisal for Suite 220 as of the ‘13 and ‘14 grand list dates?

‘‘A. That is correct.’’

The defendants’ counsel asked Leary to explain the market rental rates in his appraisal report:

‘‘Q. So, Mr. Leary, just to clarify for the record and also help for, for my understanding as well. Gross leasable area, as referenced in your appraisal, is similar to finished leasable area. Correct?

‘‘A. No. Huh, okay? The rental rate that I used—90 to 95 dollars per square foot—is a rental rate per square foot of gross leasable area. When multiplying it times the area available in the subject building, I am using— I am multiplying it by the finished leasable area, because that’s all that’s finished at this point in time. So it’s a— it’s a unit price per square foot of gross leasable area. If the other portion of the space were already fit out, it would be a higher number that I was multiplying it by. That’s all. It’s, it’s the same unit of comparison: price per square foot of gross leasable area. . . .

‘‘Q. So, Mr. Leary, we’re looking at the first page on the board, and after the expansion there was a new addition at the subject property. The gross leasable area is 182,766 square feet. Correct?

‘‘A. Will be 182,766 square feet because that includes the space that’s not yet fit out.

‘‘Q. I understand that you made a distinction between finished leasable area and unfinished leasable area. However, the 17,598 square feet in Suite 220 is area that [the plaintiff] could lease today. Correct?

‘‘A. I doubt it.’’ The court then sought the following clarification from Leary:

‘‘Q. So . . . your . . . direct capitalization approach didn’t assign a value to the unfinished space?

‘‘A. Correct. I considered it. Frankly, if you look at the space, there is only one potential use for that space. If you see Photo I-12, which will be at this point some kind of dead storage. As a matter of fact, it looks like they’re using it partially for that. There’s some old furniture and stuff in there.

‘‘So the question was, do I apply a dead storage rental to that space? But then, you have to think of the context —within a data center, with all this security and everything else like that and with raised floor going into the hallway that leads to this space, would the other tenants and would the owner want somebody trucking in and out of here with, you know, old furniture or whatever would go into a dead storage unit? The answer to me was absolutely not.

‘‘You, you—[the plaintiff] could not rent to anybody using it for dead storage and, therefore, I assigned no rent to it at this point in time.’’

At another point during trial, Leary stated that ‘‘[b]y 2013, 2014, [the plaintiff] discovered that [it] couldn’t rent [Suite 210, which it] had built out as a wholesale data center. So [it] reverted to colocation rentals, and [it] held off on building [Suite 220], because there was no demand for that data center.’’ The following colloquy then occurred between the defendants’ counsel and Leary:

‘‘Q. Mr. Leary, you’ve now, in your last few responses, talked about market demand for data center—data center space in 2013 and ‘14. Correct? ‘‘A. Yes. ‘‘Q. Okay. But you didn’t consider the market for data center space in 2013 and 2014 in your appraisal. Correct?

‘‘A. I considered the market demand in 2011. Correct.’’ Counsel for the defendants then indicated that the average rate per square foot of net rentable area for new leases rented in 2011 for other properties owned by the plaintiff was approximately $39 and asked Leary why he did not impute a market rental rate of this amount to Suite 220. Leary explained that he had no idea where the plaintiff’s other leased properties were located or what the market conditions were in those locations. Leary testified that he would need to know the location, size of the space, and the demand in the marketplace for each leased property to determine whether the rental rates were an appropriate indicator of a market rental rate for the purposes of imputing income to Suite 220. Without further information, Leary stated that he could not determine if the $39 per square foot rate was an appropriate indicator of market value for Suite 220. Leary testified, ‘‘I do know that in Trumbull , Connecticut, there’s no market for this space.’’

Lucey also testified regarding the demand for data center space in the Trumbull market. When asked if there was ‘‘demand in the market in ‘13 and ‘14 for another half [suite] of data center space in Trumbull,’’ Lucey responded, ‘‘[n]o.’’

Thus, Leary’s and Lucey’s testimony regarding the lack of a market for Suite 220 contradicts the defendants ’ claim that there is no evidence to support the court’s finding that there was no market for Suite 220 in Trumbull in 2011. See United Technologies Corp. v. East Windsor, supra, 262 Conn. 23 (‘‘[a] finding of fact is clearly erroneous when there is no evidence in the record to support it’’ (internal quotation marks omitted )). Leary’s and Lucey’s testimony plainly supports the court’s finding.

B

The defendants next argue that the court’s decision not to impute income to Suite 220 is clearly erroneous because ‘‘Suite 220 clearly adds value to the property.’’ The defendants contend that ‘‘Leary’s own cost approach analysis confirms that Suite 220 contributes value to the property.’’

The defendants’ argument rests on a faulty premise. As previously explained, the court found the income capitalization approach most reliable. Under the income capitalization approach, the value of the property is based on the property’s income producing potential , considering both actual rent and market rent. See United Technologies Corp. v. East Windsor, supra, 262 Conn. 17 n.9. Had the court concluded that another valuation method, such as the cost approach, was most appropriate, then the value of the suite could have been determined by its cost. The defendants conceded at oral argument that the income producing potential of the suite is determinative of the value of the suite under the income capitalization approach. Because there was evidence in the record that there was no actual or market rent for Suite 220, we cannot conclude that the court’s finding that no income potential existed was clearly erroneous.

C

Despite arguing that there was no evidence in the record to support the court’s finding that there was no market for Suite 220 in Trumbull in 2011, the defendants also argue that the court erred in crediting Leary’s testimony to that effect. Specifically, the defendants argue that the court erred in crediting Leary’s testimony because it was contradicted by two pieces of evidence at trial—the undisputed facts that the plaintiff invested millions of dollars in the construction of Suite 220 and actually marketed Suite 220 as leasable space. We decline to usurp the role of the trial judge.

‘‘The trier of fact must arrive at [its] own conclusions as to the value of [the taxpayer’s property] by weighing the opinion of the appraisers . . . .’’ (Internal quotation marks omitted.) Aetna Life Ins. Co. v. Middletown, supra, 77 Conn. App. 26. ‘‘It is well established that [i]n a case tried before a court, the trial judge is the sole arbiter of the credibility of the witnesses and the weight to be given specific testimony. . . . The credibility and the weight of expert testimony is judged by the same standard, and the trial court [judge] is privileged to adopt whatever testimony he reasonably believes to be credible. . . . On appeal, we do not retry the facts or pass on the credibility of witnesses.’’ (Internal quotation marks omitted.) Sakon v. Glastonbury, supra, 111 Conn. App. 252. ‘‘Because a tax appeal is heard de novo, a trial court judge is privileged to adopt whatever testimony he reasonably believes to be credible.’’ (Internal quotation marks omitted.) Grolier, Inc. v. Danbury, supra, 82 Conn. App. 79. Thus, ‘‘credibility determinations are within the exclusive province of the court.’’ Ress v. Suffield, supra, 80 Conn. App. 633.

Evidence that the plaintiff invested money in the con-

struction of the suite and that it marketed the suite was presented to the court at trial. There was also, however, testimony from Leary that there was no market for PBB space in Trumbull in 2011. In concluding that Suite 220 had no income potential, the court carefully considered all of the evidence before it and clearly credited Leary’s testimony that there was no market for PBB space in Trumbull in 2011. Although the defendants essentially invite this court to reweigh the evidence in its favor and credit Sterling’s testimony that Suite 220 was rentable and capable of generating income based on 2011 market conditions, we decline to do so.16 See Abington, LLC v. Avon, 101 Conn. App. 709, 719, 922 A.2d 1148 (2007) (‘‘[T]he determination of the credibility of expert witnesses and the weight to be accorded their testimony is within the province of the trier of facts, who is privileged to adopt whatever testimony he reasonably believes to be credible. . . . [I]t is the proper function of the court to give credence to one expert over the other.’’ (Internal quotation marks omitted.)).

D

The defendants argue that the court’s determination that there was no market for Suite 220 in Trumbull in 2011 is clearly erroneous because the court relied on post hoc information in making this determination. We reject this claim.

After concluding that Suite 220 constituted PBB space, the court stated that ‘‘[t]he more difficult question is whether the space was marketable. The fact that it was vacant does not resolve this question, for, as the town correctly points out, potential gross income comprises . . . market rent for vacant . . . space. . . . Leary testified credibly, however, that in 2011 there was no market for PBB space in Trumbull. Lucey testified, also credibly, that such space is usually rented on a building wide basis. That testimony and the economics of PBB space cast substantial doubt on whether PBB space with 16,000 rentable square feet would be marketable in the Trumbull area. In its briefs, [the plaintiff] asserts that the lack of success the last few years in renting the space or even attracting potential customers to consider it shows it to be the poster child for the poor state of the data center market in Trumbull. . . . None of the evidence about the Trumbull data center market in 2011, 2013 or 2014 credibly supports the town’s claim that Suite 220 was marketable as PBB space then.’’ (Citations omitted; emphasis added; footnote omitted; internal quotation marks omitted.)

Although the court referenced ‘‘[the plaintiff’s] assert [ion] that the lack of success the last few years in renting the space or even attracting potential customers to consider it shows it to be the poster child for the poor state of the data center market in Trumbull’’; (emphasis added; internal quotation marks omitted); the court went on to conclude that ‘‘[n]one of the evidence about the Trumbull data center market in 2011, 2013 or 2014 credibly supports the [defendants’] claim that Suite 220 was marketable as PBB space then.’’ (Emphasis added.) The defendants argue that the suite’s marketability as of October 1, 2013, and October 1, 2014, is not relevant to the determination of the suite’s marketability as of the revaluation date. The suite’s marketability as of October 1, 2013, and October 1, 2014, was relevant, however, because Sterling considered information about market conditions through October 1, 2014. Because Sterling used a cutoff date of 2014, considering information about market rent and market conditions through October 1, 2014, it was not improper for the court to consider evidence of the Trumbull data center market in 2013 or 2014.17 The defendants also argue that the court erred in considering evidence of the suite’s marketability as of the trial date and ‘‘ ‘the last few years’ ’’ prior to trial. Although the court refers to the plaintiff’s argument concerning ‘‘the lack of success the last few years in renting the space,’’ the court’s acknowledgement of this argument was not material to its decision. The court makes clear in the subsequent sentence that it specifically found that ‘‘[n]one of the evidence about the Trumbull data center market in 2011, 2013 or 2014 credibly supports the town’s claim that Suite 220 was marketable as PBB space then.’’ (Emphasis added.) Thus, the court clearly relied on evidence of the data market in Trumbull in 2011, 2013, and 2014 to make its finding that there was no market for Suite 220 as of the revaluation date.

Our case law makes clear that, ‘‘[i]n a tax appeal, the court may consider any facts that are relevant to determining whether a taxpayer actually has been overassessed .’’ (Internal quotation marks omitted.) Ress v. Suffield, supra, 80 Conn. App. 634. The court here did just that. Although the defendants take exception to the court’s consideration of the market through the cutoff date, they have not provided this court with any legal support that this was improper. We therefore conclude that the court did not err in concluding that none of the evidence credibly supported the defendants’ claim that Suite 220 was marketable as PBB space in 2011, 2013, and 2014.18

E

The defendants argue that the court’s finding that there was no market for Suite 220 is inconsistent with its market rent analysis. Specifically, the defendants argue that the court’s conclusion that the demand for PBB space was limited by the inherent characteristics of the Connecticut data center market is unfounded and contrary to its own market rent analysis. We find no inconsistency.

In completing his market rent analysis, Leary indicated that there was no specific market rent information available in the Connecticut market. Therefore, to determine what market rental rate to apply to the property, Leary applied market rental rates from the Boston data center market, finding that the Boston market was most comparable to the market in Trumbull. Leary examined five recent leases of wholesale data center spaces in the Boston area. Leary used this data, along with the data of the contract rents at the property for 2011, to determine the income potential of the property as of October 1, 2011. Leary, however, did not impute the Boston market rate to Suite 220 because he found that there was no market for that space in Trumbull in 2011.

The court found that ‘‘[t]he evidence supports and confirms Leary’s conclusion that the Trumbull data center market was comparable to the Boston one in 2011 and that comparable properties in the Boston market in the relevant time period showed a range in market rate rents from $90 to $105 per square foot of rentable space.’’ The court used this market rate in calculating the income potential of the property. The court, however , did not apply this rate to Suite 220. The court found that, although there was a national market and national demand for PBB space in 2011, there was no market for PBB space in Trumbull in 2011. The court explained that ‘‘the evidence showed that the market for this facility would lie exclusively in Connecticut based customers, and [the plaintiff] was aware of that limitation in its potential customer base. Leary’s assertion that there was no market for such space in Trumbull in 2011 is therefore found to be credible.’’

In explaining its conclusion that the market for data center space in Connecticut was limited to Connecticut based customers, the court stated that ‘‘[t]he high cost of electricity in Connecticut and the greater ‘latency’19— the delay in the amount of time it takes to send and receive electronic information between Connecticut and New York City, as compared to [that] between suburban New Jersey and New York City—limit the appeal of data centers in this state to Connecticut based customers.’’ (Footnote in original.)

The defendants argue that ‘‘[t]he suggestion that demand for PBB space was limited by the inherent characteristics of the Connecticut data center market is unfounded . . . .’’ We disagree. The court cited to evidence in the record—specifically, the plaintiff’s Investment Committee Memoranda20 and Lucey’s testimony 21—that supported its conclusion. Therefore, we conclude that the court’s conclusion that the market for the property would lie exclusively in Connecticut based customers was supported by evidence in the record.

The defendants also contend that the court’s conclusion that the demand for PBB space was limited by the inherent characteristics of the Connecticut market is contrary to its market rent analysis. The defendants claim that the court should have imputed income to Suite 220 on the basis of a lease of PBB space in Somerville , Massachusetts, signed in May, 2011, because the lease was a relevant indicator of market demand and rent for PBB space in Trumbull. We disagree.

The court’s conclusion that the demand for the property was limited by the inherent characteristics of the Connecticut market is not contrary to its market rent analysis. Leary explained at trial that the size of the potential tenants in both the Boston and Trumbull data center markets was what made them similar. Although the court found that the Trumbull market was comparable to the Boston market in that respect, the court concluded that the demand for the property was limited by the inherent characteristics of the Connecticut market , which, as previously noted, was supported by the plaintiff’s Investment Committee Memoranda and Lucey’s testimony. We therefore conclude that the court’s conclusion that the demand for PBB space was limited by the inherent characteristics of the Connecticut market is not contrary to its market rent analysis.

F

The defendants argue that the record is devoid of any evidence indicating that Suite 220’s square footage made it unmarketable as PBB space. The record reveals the contrary.

At trial, Leary testified that, ‘‘normally, this PBB . . . applies to a whole building. There are cases in which, you know, a subset of a building might be leased on a PBB basis, but, normally, you’re looking at the whole building, and it’s kind of a . . . situation where, basically , the main building [has] infrastructures in place, and then the tenant goes in and . . . does their customized finish afterward. So just, you know, the size of these things implies that, you know, it’s whole buildings. . . . [A]nd one of the things I had pointed out in my summary here was of all these large facilities, only one of them was in New England, and that was the 90,000 square foot facility . . . in Boston. Most of this occurs in other places in the market.’’

At trial, Lucey also testified to the marketability of the suite based on its square footage. The following exchange occurred between the plaintiff’s counsel and Lucey:

‘‘Q. [D]oes [the plaintiff] have [PBB space] that it rents in the northeast?

‘‘A. Yes. ‘‘Q. Give us an idea of what the typical size is of what [the plaintiff] has in the northeast?

‘‘A. Let’s see, in the northeast, probably the smallest one . . . it’s probably 50 to 60,000 square feet. The largest being [between 230,000 and] 240,000 square feet.

‘‘Q. Okay. And the, the unbuilt space in the expansion at . . . [the property], would you classify that as [PBB] space?

‘‘A. The way it’s built right now, the power is brought . . . to the shell, or we could bring the power to the shell. So . . . the short answer is, we could certainly turn that into a [PBB] shell for that 17,000 square feet. Yeah, we could do that.

‘‘Q. Do you think it’s rentable on that basis? ‘‘A. Highly unlikely. ‘‘Q. Why? ‘‘A. I, I won’t say it’s impossible, but highly unlikely. Typically, when customers rent PBB [space] . . . they’re looking for a much larger footprint. The reason for that is, because we don’t operate the data center— I mean we—we bring power to the exterior, and then the rest of it is yours. You have to staff up. You have to have people there to operate that data center. And it would be very cost ineffective for someone to have a 17,000 square foot [PBB] shell and have to have all the staffing you need to operate a data center for— there’s not enough square footage to spread the cost.’’

The defendants also introduced the plaintiff’s 2011 and 2013 Form 10-K Annual Reports, filed with the Securities and Exchange Commission (SEC), into evidence . Both documents contain the number and square footage of new and renewable PBB leases signed in those years.

In its memorandum of decision, although the court acknowledged that, theoretically, PBB space could be as small as 5000 square feet, the court found that the ‘‘testimony and the economics of PBB space cast substantial doubt on whether PBB space with 16,000 rentable square feet would be marketable in the Trumbull area.’’ The court therefore concluded that none of the evidence supported the defendants’ claim that Suite 220 was marketable in Trumbull in 2011. In making this conclusion, the court found both Lucey’s and Leary’s testimony credible. The court also relied on the plaintiff ’s 10-K financial statements for 2011 and 2013. The court explained that the average number of total square footage of new and renewal PBB leases signed supported Leary’s and Lucey’s testimony that tenants typically want larger space when renting PBB space. We conclude that, because the trial court set forth its careful consideration of the expert testimony and reports and its findings are amply supported by the record, its determination that Suite 220 had no income and no income potential in 2011 is not clearly erroneous.

II

The defendants next claim that the court’s determination that the highest and best use of Suite 210 was as wholesale space was clearly erroneous. Specifically, the defendants argue that the court’s determination of the suite’s highest and best use (1) is not supported by the evidence, (2) is inconsistent with the court’s determination that there was no market for PBB space in Trumbull in 2011 due to the inherent limitations of the Connecticut based data center market, and (3) fails to take into account the plaintiff’s existing leases and actual rent generated by those leases in its market rent analysis. We disagree.

Suite 210 is located in the expansion of the property and is a fully built out data center suite. After the expansion was completed, the plaintiff was unable to find a wholesale tenant to lease Suite 210 as it had hoped. As a result, the plaintiff began leasing the suite to colocation tenants in mid-2013. By 2014, 41 percent of Suite 210’s raised floor space was leased to seven colocation tenants . The average rent paid by these tenants in 2014 was $404 per square foot.

In their appraisal reports and at trial, Sterling and Leary disagreed on the highest and best use of Suite 210 and, therefore, on what market rental rate to apply to calculate the suite’s income potential. Sterling was of the opinion that, as of the date of valuation, the highest and best use of Suite 210 was as colocation space. Leary, on the other hand, concluded that the highest and best use of Suite 210 was as wholesale space.

In its memorandum of decision, the court concluded that the highest and best use of Suite 2010 in 2011 was as wholesale space and, accordingly, determined that market rent for Suite 210 properly should be regarded as based on wholesale data rates in 2011. In reaching this conclusion, the court first noted that the fact that the plaintiff was not leasing Suite 210 as colocation space in 2011 and the plaintiff’s intention to lease the suite for wholesale use was not conclusive of the suite’s highest and best use: ‘‘The fact that [the plaintiff] itself did not begin marketing colocation space until 2012 does not preclude colocation from being the highest and best use of Suite 210 in 2011.’’ The court did not find credible Leary’s opinion that there was no market for colocation space in Trumbull in 2011. Rather, the court found credible Sterling’s description of the colocation space as an evolving ‘‘ ‘sweet spot’ ’’ of the data center market. The court credited Sterling’s testimony that colocation was becoming more popular. The court determined, however, that Sterling’s testimony did not preclude the court from agreeing with Leary’s conclusion that the highest and best use of the suite was as wholesale space, especially in light of Lucey’s testimony . The court found Lucey’s testimony to be credible, particularly with regard to his ‘‘catalogue of reasons that colocation rentals are less desirable than wholesale rentals . . . .’’ This catalogue, according to the court, included the fact that colocation leases tend to involve (1) a shorter duration, (2) smaller amounts of power, (3) tenants who occupy less space, (4) increased risk due to tenants who are typically less creditworthy, (5) greater periods of time between leases, (6) higher transactional costs due to higher turnover rates, and (7) flat rates based on the amount of kilowatts provided to tenants rather than both such flat rates and the pro rata add-ons paid by wholesale customers for their share of certain building and common area expenses and electricity. The court found that the plaintiff’s seven colocation leases in 2013 and 2014 shared some of these characteristics. The court also credited Lucey’s testimony that the difference in rental market rates between colocation and wholesale customers is ‘‘ ‘not as great as people think.’ ’’

On the basis of our review of the record, we have little difficulty concluding that the court’s finding was supported by evidence in the record. On direct examination , Lucey explained why colocation leases were less desirable to the plaintiff than wholesale leases: ‘‘[T]hey were unable to lease [Suite 210] for wholesale. So they had to go into a [colocation] leasing arrangement where you have multiple tenants, which . . . it’s typically a longer lease-up process. It also tends to be much more hands-on and, and resource demanding just because . . . people are leasing very small bits of power and space. You also have the transactional cost and the transactional volume increases. You have to do many, many more leases, and your lease-up time tends to be extended, and you don’t typically get long terms.’’

During cross-examination, Lucey explained that, despite the perceived price difference between colocation and wholesale leases, the actual difference is not very significant: ‘‘Generally speaking, I would say . . . that a [colocation] customer would pay . . . a premium to . . . a wholesale customer. . . . The reasons are . . . [colocation] customers tend to . . . lease [a] smaller amount . . . . [Colocation] leases typically are shorter. So you charge a premium, because you’re not getting the term. You also tend to not get the same credit quality. So you try to charge a premium for that. In addition to that, they’re just leasing much less. So you would tend to try to charge a premium for that. Now, what’s offset in the [colocation] world is your transaction costs are much higher because you’re turning over the space far more often. So you’ve got more downtime baked into it. In addition to that you’re paying more brokerage commissions. . . . [W]hen you lease [colocation] space, it’s typically done on an all-in basis. . . . [Y]ou’re not going to see another bill from us for electricity or anything else. . . . Wholesale is . . . typically priced differently. . . . [O]n wholesale . . . you pay a base rate. So maybe you’re paying 140 a kilowatt versus the [colocation] customers [who are] paying 200 a kilowatt. But as the wholesale customer, you’re also being charged a pro rata share of a building’s expenses, common area expenses, like . . . janitorial, plowing, all of that stuff. You pay a pro rata share of that. In addition to that, you also pay a separate charge for any electricity you use in your space above and beyond the contracted-for power. . . . Light bulbs. You run computers. People have offices, command centers . . . . [T]here are a number of additional add-ons to the . . . wholesale person that don’t apply to [colocation ]. . . . [S]o there is a price difference. It’s not as great as people think. . . . Yes . . . you definitely charge a premium, but when you . . . strip it all back, the . . . differences aren’t that great . . . .’’

This evidence was sufficient to support the court’s finding. Nevertheless, the defendants argue that the court erred by failing to consider the plaintiff’s existing leases and the actual rent from those leases in its market rent analysis. The defendants contend that Suite 210 was devoted to the use for which it was best adapted— colocation—and, therefore, the court erred by not applying the actual rent from the plaintiff’s 2013 and 2014 colocation leases to determine the suite’s income potential.

As previously explained, ‘‘[p]ursuant to § 12-63b (b), the court is required to consider both market rent and actual rent when determining fair market value using the income capitalization method. . . . [I]f the property is devoted to the use for which it is best adapted and is in a condition to produce or is producing its maximum income, the actual rental is a very important element in ascertaining its value.’’ (Citation omitted; footnotes omitted; internal quotation marks omitted.) Pilot’s Point Marina, Inc. v. Westbrook, 119 Conn. App. 600, 603–604, 988 A.2d 897 (2010). Here, on the basis of the evidence in the record, the court concluded that the suite was not devoted to the use for which it was best adapted. Although the suite was being leased as colocation space in 2013 and 2014, the court concluded that the suite’s highest and best use was to be leased as wholesale space. Contrary to the defendants’ assertion, the court did consider the plaintiff’s existing leases and actual rent from those leases. Because those leases were for colocation customers, however, the court did not use the actual rent from those leases to determine the suite’s fair market value. Rather, the court properly applied market rental rates to determine the suite’s fair market value as wholesale space. We therefore conclude that the court properly considered both the suite’s actual rent and market rent. See id.

The defendants also contend that the court’s determination of the highest and best use of Suite 210 is inconsistent with its determination that there was no market for PBB space due to the inherent limitations of the Connecticut based data center market. In concluding that there was no market for PBB space in Trumbull in 2011, the court found that the demand for such space would lie exclusively with Connecticut based customers because these customers often had more localized demand. The court also noted that the customers were often smaller users. The court found Lucey’s testimony that PBB space is usually rented on a building wide basis credible. Although the court found that Connecticut based customers tend to be smaller users with more localized demand, the court nevertheless concluded that 16,000 rentable square feet of PBB space was not marketable to those customers. The court’s conclusion that there was no market for PBB space in Trumbull in 2011 does not contradict its conclusion that the highest and best use of Suite 210, a fully built out suite, would be wholesale space. The court’s conclusion that there was no demand for PBB space in Trumbull in 2011 does not preclude it from also finding that there was demand for a wholesale lease of a fully built out suite. We therefore conclude that the court’s determinations are not contradictory. On the basis of the evidence in the record, we conclude that the court’s determination that wholesale use was the highest and best use of Suite 210 in 2011 was not clearly erroneous.

III

The defendants next argue that the court erred in applying a capitalization rate of 8 percent. Specifically, the defendants contend that the court applied an ‘‘unsubstantiated’’ capitalization rate.22 The last two steps in the income capitalization approach include selecting an applicable capitalization rate and applying that capitalization rate to the property ’s net income to arrive at an indication of the market value of the property being assessed. See Sun Valley Camping Cooperative, Inc. v. Stafford, supra, 94 Conn. App. 702 n.9. ‘‘The capitalization rate is considered the rate a reasonable investor would seek on his capital or equity . . . . In other words, the capitalization rate is a projection of the buyer’s return on investment based upon a comparison of the property’s income producing capacity to its purchase price.’’ (Citation omitted; internal quotation marks omitted.) Fairfield Merrittview Ltd. Partnership v. Norwalk, supra, 172 Conn. App. 166 n.14. ‘‘It is generally recognized that the most difficult element to determine, and so far as results are concerned the most important, in any computation of value based on earning power is the rate at which the earnings are to be capitalized.’’ Burritt Mutual Savings Bank of New Britain v. New Britain, 146 Conn. 669, 676, 154 A.2d 608 (1959). ‘‘Determination of the capitalization rate is both critical and highly subjective.’’ 1 Powell on Real Property (M. Wolf ed., 2017) § 10B.06 [4] [c] [v], p. 10B-111.

The two appraisers, Sterling and Leary, used different methodologies for selecting an applicable capitalization rate. Sterling used the band of investment technique.23 The band of investment technique recognizes the presence of both the debt and equity components of commercial real estate. See Appraisal Institute, supra, p. 535 (explaining that band of investment technique is ‘‘[a] technique in which the capitalization rates attributed to components of a capital investment are weighted and combined to derive a weighted-average rate attributable to the total investment’’). Using this technique, Sterling applied a capitalization rate of 6.91 percent.24 Leary reached his capitalization rate by analyzing sales of like properties and reviewing recognized market surveys of investors in such properties. See Appraisal Institute, supra, p. 530 (‘‘[o]verall capitalization rates can be estimated with various techniques’’); id., p. 530 n.2 (‘‘[s]urveys of overall capitalization rates based on the market expectations of lenders and owners are available, but such data should be rigorously scrutinized’’). Leary analyzed information on capitalization rates of like properties in the United States with similar types of income streams. Specifically, he reviewed certain newsletters from professional real estate organizations that focused on data centers, which included published capitalization rates from the sale of seven data centers across the United States. Using this information, Leary concluded that an appropriate capitalization rate for the property was between 8.25 and 8.5 percent.25 The court noted that, although both techniques were commonly used methods for determining an appropriate capitalization rate, ‘‘[t]here [were] some concerns about each appraiser’s methodology.’’ The court ultimately concluded that the appropriate capitalization rate was 8 percent, finding that Leary’s capitalization rate was ‘‘essentially correct, although it [did] not give enough consideration to the effect of the brand new electrical infrastructure installed with the remediation and the expansion.’’

The defendants contend that the court’s finding that 8 percent represented an appropriate capitalization rate was clearly erroneous. The defendants take issue with the court’s capitalization rate because, in their view, the newsletters and surveys Leary relied on and, therefore , on which the court relied in finding Leary’s capitalization rate ‘‘ ‘essentially correct,’ ’’ rely on ‘‘unsubstantiated and unverified market data.’’ Specifically, the defendants argue that Leary could not verify certain information about the seven data centers in the market newsletters, including (1) their square footage, (2) the square footage of their raised floor area, (3) their electrical capacity, (4) their age, (5) their condition, and (6) their income and expenses. Thus, the defendants argue that Leary could not confirm whether the data centers referenced in the newsletters were comparable to the property, and it was clearly erroneous for the court to rely on such newsletters in determining a proper capitalization rate for the property. We disagree.

Where ‘‘the trial court is confronted with conflicting accounting methods . . . giving credence to one over the other is a proper exercise of its function as a trier of fact.’’ Connecticut Coke Co. v. New Haven, 169 Conn. 663, 666, 364 A.2d 178 (1975). ‘‘In arriving at the value of property, no one method is controlling, and there is no rule of law that any particular method of valuation must be followed. It is a matter of opinion based on all the evidence and, at best, is one of approximation.’’ (Internal quotation marks omitted.) Housing Authority v. CB Alexander Real Estate, LLC, 107 Conn. App. 167, 180, 944 A.2d 1010 (2008). ‘‘In reviewing valuations, we must bear in mind that the process of estimating the value of property for taxation is, at best, one of approximation and judgment, and that there is a margin for a difference of opinion.’’ (Internal quotation marks omitted .) Connecticut Coke Co. v. New Haven, supra, 668. ‘‘We will disturb the trial court’s determination of valuation , therefore, only when it appears on the record before us that the court misapplied or overlooked, or gave a wrong or improper effect to, any test or consideration which it was [its] duty to regard.’’ (Internal quotation marks omitted.) New Haven Savings Bank v. West Haven Sound Development, 190 Conn. 60, 70, 459 A.2d 999 (1983).

Here, the court properly gave credence to one appraiser’s method of calculating the capitalization rate over the other and determined an appropriate capitalization rate based on evidence in the record. The record includes excerpts from the Avision Young Data Center Practice newsletter, which reported seven transactions referencing capitalization rates for those sales.26 The newsletter reported an overall capitalization rate range from 6.2 percent to 10.2 percent. Leary testified that the two lowest rates involved one purchaser who bought two properties in California and Virginia ‘‘at the same time . . . in very active markets.’’ Leary testified that he believed the investor ‘‘clearly had some investment criteria that were different than the majority of the transactions’’ and those transactions ‘‘appear to be below market based on the remaining five transactions .’’ The remaining five transactions reported in the newsletter had capitalization rates ranging from 8.1 percent to 10.2 percent. Leary testified that the two lowest capitalization rates from these five transactions were located in Georgia, which has lower utility costs than Connecticut and less risk, resulting in a lower capitalization rate. The transaction with the highest capitalization rate was located in Michigan, which Leary testified has higher utility costs.

The record also included a Real Estate Research Corporation Quarterly Real Estate Report (RERC report).

Both Leary and Sterling referenced the RERC report in their appraisal reports.27 The RERC report includes capitalization rates for different types of properties in the eastern United States within different categories including, inter alia, ‘‘[o]ffice,’’ ‘‘[w]arehouse,’’ and research and development.28 The capitalization rate ranges for first tier29 properties were 5.5 percent to 10 percent with an average of 7.9 percent for suburban offices and 5 percent to 12 percent with an average of 8.2 percent for research and development properties.30 The capitalization rate for second tier research and development properties ranged from 6 percent to 10.5 percent with an average of 8.6 percent.

In reaching its conclusion that 8 percent was an appropriate capitalization rate, the court stated that the property had characteristics of both a first tier and a second tier property. The court explained that the expansion and electrical remediation were characteristics of a first tier property, while the location in Trumbull was a characteristic of a second tier property. The court found that Sterling too easily dismissed the threat of obsolescence in the data center industry, while Leary’s classification of the property as a second tier property recognized the potential for obsolescence. The court concluded that Leary’s capitalization rate was essentially correct, although his capitalization rate did not give enough consideration to the effect of the brand new electrical infrastructure installed with the remediation and the expansion. The court thus concluded that the appropriate capitalization rate, properly taking the new electrical infrastructure into consideration, was 8 percent, slightly lower than Leary’s range of 8.25 percent to 8.5 percent. Because the court’s capitalization rate was supported by evidence in the record, we conclude that its finding was not clearly erroneous.

IV

The defendants’ final claim is that the court’s determination of the fair market value of the property is clearly erroneous because the court disregarded the plaintiff’s internal valuations, which clearly undermine the court’s valuation of the property. The defendants point to the plaintiff’s 10-K Forms for the 2013 and 2014 years, in which the plaintiff reported the property and 80 Merritt Boulevard as having a combined book value of $163,486,000 and $165,596,000 for 2013 and 2014, respectively. The defendant also relies on an asset ‘‘impairment analysis’’ completed by the plaintiff in 2014, which the defendant argues shows that the value of the property was $180,293,972 as of 2014, based on market conditions as of 2011.

In its Form 10-K Annual Reports submitted to the SEC, the plaintiff reported that the combined book value of the property and 80 Merritt Boulevard was $163,486,000 in 2013 and $165,596,000 in 2014. The defendants contend that the court erred in disregarding these book values in determining the property’s fair market value. Lucey, however, testified that the values reported to the SEC in the 10-K Forms represented the book value, rather than the fair market value, of the property.31 As the trial court explained in its memorandum of decision, the plaintiff’s 10-K filings show ‘‘ ‘book value’, not fair market value, and have little relation to the issues here.’’

The defendants also contend that the asset ‘‘impairment analysis’’ completed by the plaintiff in 2014 shows that the value of the property was $180,293,972 as of 2014, based on market conditions as of 2011, undermining the court’s valuation of the property. The defendants rely on a document created by the plaintiff as part of this analysis, which states that, on the basis of the ten year cash flow of the property and 80 Merritt Boulevard, the properties were valued at $198,200,119 in 2014 and $212,744,378 in 2015.

As previously described, the plaintiff’s projected returns on its investment into the property did not come to fruition. The property was included in a 2014 review of potentially underperforming properties. The plaintiff ’s accounting group then undertook an ‘‘ ‘impairment analysis.’ ’’ As part of the impairment analysis, the plaintiff completed a recoverability test. This test involved a ‘‘dynamic valuation’’ of both the property and 80 Merritt Boulevard and included a ten year cash flow analysis based on certain optimistic assumptions. The plaintiff’s internal documents show that the ten year dynamic valuation of cash flow of the property and 80 Merritt Boulevard was $198,200,119 in 2014 and $212,744,378 in 2015. The defendants argue that subtracting the parties’ stipulated value for 80 Merritt Boulevard results in a value of $186,371,800 for the property for 2014. After applying the 3 percent discount factor used by Leary, the defendants contend that the value of the property was $180,293,972.

At trial, however, Lucey described the ten year dynamic valuation of cash flow analysis as a ‘‘best case scenario . . . .’’ The plaintiff describes the process as an ‘‘optimistic projection, establishing the upper limit of [return on invested capital].’’ The analysis did not take into account variables such as leasing pace, structural vacancy, structural free rent, carrying costs, new development capital, or portfolio changes during lease- up of vacant space. The analysis also assumed 95 percent occupancy by 2017. Also as part of this analysis, the plaintiff estimated the combined fair market value of the property and 80 Merritt Boulevard to be $120,000,000. At trial, Lucey explained that this fair market value included not only both properties but also personal property, such as equipment on site.

In its memorandum of decision, the trial court stated that the ‘‘$120 million estimation of fair market value used during the 2014 Return on Invested Capital analysis was placed on the combined properties and personalty within and, together with the parties’ stipulations about the fair market value of 80 Merritt Boulevard, is consistent with this court’s valuation for [the property ].’’ The court reasoned that the ten year dynamic valuation of cash flow was based on optimistic assumptions and a best case scenario. Thus, the court clearly considered but did not place weight on the plaintiff’s internal valuations.

After reviewing the record, including the thorough and well reasoned memorandum of decision, we conclude that the court’s determination of the fair market value of the property was not clearly erroneous.

The judgments are affirmed. In this opinion the other judges concurred.

1 The plaintiff filed two separate appeals with the Superior Court in the judicial district of Fairfield pursuant to General Statutes § 12-117a. In the first appeal, Docket No. CV-XX-XXXXXXX-S, the plaintiff appealed from the 2013 valuation of the board, which valued 60 and 80 Merritt Boulevard at a true and actual value of $145,446,143. In the second appeal, Docket No. CV-XX-XXXXXXX-S, the plaintiff appealed from the 2014 valuation of the board, which valued 60 and 80 Merritt Boulevard at a true and actual value of $145,446,145. The appeals were consolidated on May 9, 2014, and transferred to the judicial district of New Britain.

2 The plaintiff is a subsidiary of Digital Realty Trust, Inc., which is a publicly traded company. Digital Realty Trust, Inc., as its name suggests, is a real estate investment trust, formed in 2004.

3 The court issued its memorandum of decision on August 4, 2020. The plaintiff subsequently filed a motion to correct the memorandum of decision to correct ‘‘five minor errors.’’ The defendants objected to the second of the five requested corrections. The plaintiff then withdrew the second requested correction. The court granted the first, third, and fifth requested corrections and granted the fourth in part. On February 16, 2021, the court issued its corrected memorandum of decision, correcting the scrivener’s errors in the original decision.

4 At trial, the parties’ experts disagreed as to the precise square footage of the building. The plaintiff’s expert, John Leary, concluded that the square footage of the building was 204,024 square feet while the defendants’ expert, Russell Sterling, reached a conclusion of 193,650 square feet. The court determined that Leary’s conclusion was more reliable and credible. The property’s square footage is not an issue on appeal.

5 For example, as a result of the power outage, one tenant was given a 20 percent monthly rent abatement of $22,869.

6 A grand list is a listing of all taxable property located within a town as of the relevant tax year. See General Statutes § 12-55 (a).

7 The town had conducted the previous town wide revaluation in 2005, meaning the next town wide revaluation was to occur in 2010. That revaluation , however, was delayed one year from 2010 to 2011. In its 2011 town wide revaluation, the assessor of the town concluded that the fair market value of the property was $66,124,600. The plaintiff appealed this assessment, and the parties subsequently stipulated to a fair market value of $62,450,000. The next revaluation occurred in 2015, in accordance with the regular five year schedule.

8 Market value is ‘‘[t]he most probable price, as of a specified date, in cash, or in terms equivalent to cash, or in other precisely revealed terms, for which the specified property rights should sell after reasonable exposure in a competitive market under all conditions requisite to a fair sale, with the buyer and seller each acting prudently, knowledgeably, and for selfinterest , and assuming that neither is under undue duress.’’ Appraisal Institute , The Appraisal of Real Estate (12th Ed. 2001) p. 22.

9 The assessor concluded that the assessed value was $101,812,300. Pursu-

ant to General Statutes § 12-62a, the assessed value of a property is 70 percent of the appraised/fair market value.

10 The Appraisal Standards Board’s Advisory Opinion 34 states: ‘‘A retrospective appraisal is complicated by the fact that the appraiser already knows what occurred in the market after the effective date of the appraisal. Data subsequent to the effective date may be considered in developing a retrospective value as a confirmation of trends that would reasonably be considered by a buyer or seller as of that date. The appraiser should determine a logical cut-off for the data to be used in the analysis because at some point distant from the effective date, the subsequent data will no longer provide an accurate representation of market conditions as of the effective date. This is a difficult determination to make. Studying the market conditions as of the date of the appraisal assists the appraiser in judging where he or she should make this cut-off. With market evidence that data subsequent to the effective date was consistent with market expectations as of the effective date, the subsequent data should be used. In the absence of such evidence, the effective date should be used as the cut-off date for data considered by the appraiser.’’ Appraisal Standards Board, Appraisal Foundation, 2016–2017 Uniform Standards of Professional Appraisal Practice (2016) p. 194.

11 A fee simple interest means ‘‘absolute ownership unencumbered by any other interest or estate . . . .’’ Appraisal Institute, supra, p. 68; see also Frank Towers Corp. v. Laviana, 140 Conn. 45, 52, 97 A.2d 567 (1953) (fee simple interest means ‘‘whole or unlimited estate’’).

12 The court exercised its discretion not to award interest to the plaintiff, explaining that ‘‘this is a complex property presenting many issues affecting valuation.’’

13 At oral argument, the defendants stated that the only claim for which plenary review would be appropriate is its claim that the court violated Advisory Opinion 34, in that it considered evidence after the established cutoff date. See Appraisal Standards Board, Appraisal Foundation, 2016– 2017 Uniform Standards of Professional Appraisal Practice (2016), p. 194. This claim challenges the implementation of a particular appraisal method— the income capitalization approach—rather than the propriety of the appraisal method and, thus, does not ‘‘[reject] a method of appraisal as a matter of law . . . .’’ (Emphasis added; internal quotation marks omitted.) Walgreen Eastern Co. v. West Hartford, supra, 329 Conn. 494. Therefore, we decline to apply plenary review to this claim.

14 For the sake of clarity and ease of discussion, we address the defendants’

arguments in a different order from which they were briefed.

15 Specifically, Leary testified that the use of Suite 220 as storage space ‘‘would violate all the conditions associated with the confidentiality of all the data center tenants . . . .’’

16 The defendants also argue that Leary did not provide any support for his assertion that there was no market for Suite 220 in Trumbull in 2011. The defendants have not preserved this claim for appellate review. The defendants did not object to Leary’s testimony that there was no market for Suite 220 in Trumbull in 2011 at trial. See State v. Crocker, 83 Conn. App. 615, 653, 852 A.2d 762 (when ‘‘defendant never objected to the court’s admitting [the witness’] testimony into evidence,’’ defendant’s claim that it was improper for court to admit witness’ testimony was unpreserved), cert. denied, 271 Conn. 910, 859 A.2d 571 (2004). Accordingly, we decline to review the defendants’ unpreserved claim.

17 See footnote 10 of this opinion. 18 The defendants also contend that Leary impermissibly relied on post hoc information by testifying that, ‘‘had there [been] a market . . . then Suite 220 might well have been rented by now.’’ According to the defendants, Leary’s reliance on what occurred between 2013 and 2019 was antithetical to the Appraisal Standards Board’s Advisory Opinion 34; see Appraisal Standards Board, Appraisal Foundation, 2016–2017 Uniform Standards of Professional Appraisal Practice (2016), p. 194; see also footnote 10 of this opinion; and his cutoff date of December 31, 2012. We conclude that the defendants have not preserved this claim for appellate review, as the defendants did not object to Leary’s testimony at trial. See State v. Crocker, 83 Conn. App. 615, 653, 852 A.2d 762 (when ‘‘defendant never objected to the court’s admitting [the witness’] testimony into evidence,’’ defendant’s claim that it was improper for court to admit witness’ testimony was unpreserved), cert. denied, 271 Conn. 910, 859 A.2d 571 (2004). Accordingly, we decline to review the defendants’ unpreserved claim.

19 ‘‘Sterling’s appraisal explained the importance of latency for data center operations: The highest percentage of users of data center space are entities that . . . require transaction processing and eCommerce. Transaction processing includes financial services firms using high speed trading that require low latency so trades can be processed in milliseconds. Latency, simply defined, is the delay in the amount of time it takes to send and receive electronic information. For capital markets speed is essential since markets are volatile. In addition, traders want to be able to execute trades faster than their competitors. Physical distance from the servers as well as other factors such as bandwidth through the internet connections affect the speed of a transmission. So, the closer the brokerage offices are to the financial exchanges, the lower the latency.’’ (Internal quotation marks omitted.)

20 The plaintiff’s Investment Committee Memoranda provide in relevant part: ‘‘This facility will be positioned to attract financial tenants from Fairfield County. . . . Sales report demand of approximately 2.0 MW from other financial investment companies who have future requirements commencing in the next 24 months in the Trumbull area.’’

‘‘The Connecticut sect of the Metro-NY market . . . operates outside the parameters of the NJ datacenter market, which provides low latency relief for businesses operating in the high rent and utility districts of Manhattan. Businesses that choose to locate their critical infrastructure in Connecticut have an implicit desire to be there, and in many instances cannot locate their infrastructure in NJ either because of application latency issues or tariff issues.

‘‘The Connecticut data center market is characterized by uncertainties surrounding utility infrastructure and the high cost of power. This is evidenced by the limited number of datacenter providers within the state.

***

‘‘Market research and customer sentiment regarding the Trumbull, CT area indicates that a leasing strategy targeting local businesses will yield better results when compared to offering this facility as an alternative to NY/NJ market offerings. The high total cost of occupancy in the CT market creates a demand profile that requires customers to locate there.’’

21 The court stated that it found the following testimony from Lucey to be credible: ‘‘[T]he size of the market in the greater Trumbull, Connecticut area and the size of the Boston market is very similar. New Jersey is very different. It’s got a different customer base. It’s got a different—a much different size market. The New Jersey market is probably three times the size of either the Boston or the Trumbull markets, and it has a very different user profile. There are financial service firms that, that do contract with us as, as, of course, you know. Those firms, though, tend to be, um—frankly, they don’t tend to do it much anymore—but those that do, they tend to be—and this is the same— the same is true in Boston. The financial service firms—and, and our customers generally in both the Trumbull and Boston markets are typically local. They’re people that are in the Boston market or within a certain—you know, probably a relatively narrow radius of either the Boston data center or the Trumbull data center.’’

22 To the extent the defendants challenge Leary’s appraisal report or his testimony regarding the appropriate capitalization rate, we decline to review this claim. The defendants failed to object to the admission of Leary’s appraisal report containing calculations of his capitalization rate and likewise failed to object to his testimony concerning these calculations. See State v. Crocker, 83 Conn. App. 615, 653, 852 A.2d 762 (when ‘‘defendant never objected to the court’s admitting [the witness’] testimony into evidence,’’ defendant’s claim that it was improper for court to admit witness’ testimony was unpreserved), cert. denied, 271 Conn. 910, 859 A.2d 571 (2004). Accordingly , we decline to review the defendants’ unpreserved claim.

23 The defendants argue that, ‘‘[g]iven the lack of information available on data center sales, the appropriate technique for calculating the property’s capitalization rate was the band of investment technique utilized by the [defendants’] appraiser [Sterling].’’ At oral argument, when asked to which claims plenary review applies, the defendants did not refer to this claim.

We conclude that plenary review does not apply to this claim. The court did not reject either appraiser’s approach to calculating the appropriate capitalization rate as a generally applicable rule of law. See Walgreen Eastern Co. v. West Hartford, supra, 329 Conn. 493–94. We also conclude that the abuse of discretion standard does not apply to this claim because the court did not conclude that either ‘‘appraiser’s calculations were incorrect or based on a flawed formula in [the] case, or because it determined that an appraisal method was inappropriate’’ for the property. Id. Rather, the court was confronted with conflicting methods and calculations of the appro-

priate capitalization rate and properly gave credence to one over the other as the trier of fact. See Connecticut Coke Co. v. New Haven, 169 Conn. 663, 666, 364 A.2d 178 (1975); see also Abington, LLC v. Avon, supra, 101 Conn. App. 719. Thus, we apply the clearly erroneous standard of review in reviewing this claim.

24 The court explained how Sterling employed the band of investment technique to reach a capitalization rate of 6.91 percent: ‘‘He described his first step as determin[ing] . . . an appropriate rate . . . for the mortgage component of that overall capitalization rate. . . . To do so, he examined investment bulletins and surveys from the life insurance industry, RERC, Price Waterhouse Coopers Korpacz (PwC), and the mortgage lending industry , along with yield on the monthly average for ten year treasury bills. Using this data, he determined that there is support for a mortgage with an interest of 5.25 percent as of October 1, 2011, which he mathematically converted to a mortgage constant of 7.19 percent. . . . For the return on equity component, he again focused on his conclusion that the space can be marketed to the highest quality tenants increasing the quantity, quality and durability of the future income . . . and considered short-term treasury bills, corporate bonds, and extrapolated equity dividend rates from the life insurance tables he used. . . . He had previously opined in his 2011 appraisal that a return on equity rate of 8 percent was appropriate for the property. . . . For the current appraisal, he said that he decided that an equity dividend rate of 6.5 percent is appropriate because of the installation of new electrical infrastructure for the remediation, the construction of a new 70,000 square foot addition, and [the plaintiff’s] decision to begin marketing to colocation tenants. . . . Blending these two components, he arrived at a capitalization rate of 6.91 percent, which he determined was appropriate after comparing reported capitalization rates for the highest value properties, such as large offices and large industrial investments.’’ (Citations omitted; emphasis omitted; footnotes omitted; internal quotation marks omitted.)

25 The court explained how Leary determined that a capitalization rate between 8.25 and 8.5 percent was appropriate: ‘‘Leary began by examining excerpts from certain newsletters from professional organizations in the real estate business that focused on data center information. . . . Those newsletters reported sales of seven data centers in the United States during 2011 and 2012 that displayed an overall capitalization rate . . . range of 6.2 percent to 10.2 percent. . . . The two lowest rates involved one purchaser who bought two properties in California and Virginia at the same time . . . in very active markets. . . . Leary said he believed this investor clearly had some investment criteria that were different than the majority of the transactions . . . and that those purchases, in Leary’s opinion, appear to be below market based on the remaining five transactions. . . . The remaining five transactions display cap rates that range from 8.1 percent to 10.2 percent and average 8.8 percent. . . . The next two lowest capitalization rates were in Georgia, which he said has lower utility costs than Connecticut , and, thus, less risk, [and] a lower cap rate . . . and the highest was in Michigan, which has higher costs. He also considered investor reports from the Real Estate Research Corporation (RERC) Quarterly Real Estate Report. Although the RERC information did not contain data about data centers, he decided the industrial/R&D category, industrial/research and development was the closest of these categories to what might being considered a data center. . . . So I used that category, and I looked at the eastern U.S. market for the third quarter of 2011, and it showed a range of 6 percent to 10.5 percent in the market, which, ironically, was pretty close to the range of the data from the sales, and it showed an average of 8.6 percent. . . . He then testified: Well, I used both sets of data to conclude that the appropriate capitalization rate for this property, given the fact a third of the building is new, would be slightly below that average or 8.25 to 8.5 percent as the basis for converting that income into value.’’ (Citations omitted ; internal quotation marks omitted.)

26 Leary summarized the data on these seven sales in the following table:

Location Building Square Date Capitalization Footage Rate Rancho Cordova, CA 69,000 2011 9.0% Atlanta, GA 338,000 2011 8.2% Norcross, GA 33,000 2012 8.1% Southfield, MI 53,000 2012 10.2%

Austin, TX 62,000 2012 8.5% Vienna, VA 225,038 2012 6.2% San Diego, CA 166,892 2012 7.5% 27 Notably, the RERC report was admitted into evidence as part of Sterling’s appraisal report.

28 Leary testified that he concluded the research and development category to be the most applicable to the property.

29 The court explained the difference between first tier and second tier properties: ‘‘[F]irst tier refers to new or newer quality construction in prime to good locations; second tier refers to aging, formerly first tier properties in good to average locations.’’ (Internal quotation marks omitted.)

30 Leary summarized the capitalization rates reported in the RERC report for first tier properties in the eastern United States:

Regional - Eastern: Suburban Office Research and 1st Tier Development Range 5.5%–10% 5%–12% Average 7.9% 8.2% 31 When asked to explain what the $165,596,000 value reported in the SEC form 10-K for 2014 represented, Lucey testified, ‘‘[F]or our reporting purposes to the SEC, we report on book value of our assets, which essentially means what we’ve invested in the asset.’’ Lucey testified that the $165,596,000 did not represent the fair market value of the property because the plaintiff ‘‘report[s] on book value. So this is just based on what we’ve invested in the property, not what the property would sell for. . . . The value of that property—the fair market value, what someone would pay, could be dramatically higher or dramatically lower than that number. It depends on a whole host of conditions that have very little to do with what we’ve invested in the property.’’