The best test of property value is market sales. If there have been no comparable sales for many years, other tests must be employed, no single one being controlling. The process of valuation is at best one of approximation and judgment, and there is a margin for a difference of opinion. The assessment of property for taxation is an administrative proceeding, and a wide discretion is accorded the assessors. A taxpayer who thinks he is aggrieved by the refusal of a board of tax review to reduce the valuation by assessors has an appeal to the courts, where the matter of value may be tried de novo. The contesting taxpayer, however, has the burden of establishing that the assessment is unjust and will result in his being required, in effect, to pay an illegal tax. If the case is one where value must be determined largely by the use of a mathematical process, such as the capitalization of stabilized net income, the burden on the taxpayer, on appeal from the board of tax review, requires that he establish that the factors employed by the assessors in the mathematical process were improper and resulted in an illegal assessment. Unless he does establish this, the court will have no occasion to exercise the broad discretionary power which the statute confers on it for the correction of the assessment. The assessors determined the value of the plaintiff's land by the land residual method: This method, which the trial court found was a proper one to use, involved capitalization of the income regarded as attributable to the land in this so-called 100 per cent business district. One vital factor in the method, as in any computation of value based on earning power, was the rate at which the earnings were to be capitalized. Of importance also, since the earnings capitalized were found by deducting from the gross rents the income and expenses attributable to the building, was the formula used for amortization of building costs. The impropriety found by the trial court in the assessors' computations was their use of a 5.5 per cent capitalization rate and their employment of a sinking fund method of amortization. The rejection by the court of this capitalization rate was based on a finding that mortgage money could not be had at as low an interest rate as the assessors assumed. The rejection by the court of the sinking fund method of amortization was based on a finding that the method prevailing in New Britain for amortizing the capital investment in real estate such as that of the plaintiff was the so-called straight line method: Since neither of these findings was supported by evidence, the judgment reducing the assessment could not stand. The true value of a fixed asset such as real estate is fairly constant and must be gauged by conditions prevailing over a period of time. One vice in the plaintiff's proof of interest rates was that it was limited to those of a single date, the assessment date.
Argued June 4, 1959
Decided July 31, 1959
Application in the nature of an appeal from the action of the board of tax review of the defendant city in refusing to reduce the valuation of the plaintiff's real estate, brought to the Court of Common Pleas in Hartford County and tried to the court, Klau, J.; judgment sustaining the appeal and reducing the valuation and assessment, from which the defendant appealed to this court. Error; new trial.
The appellee filed a motion for reargument which was denied.
Valentine J. Sacco, with whom were Jerome I. Walsh and, on the brief, George J. Coyle, corporation counsel, for the appellant (defendant).
Harold J. Eisenberg, for the appellee (plaintiff).
In the city of New Britain the tax assessment date is September 1 in each year. On September 1, 1956, the board of assessors of the city assessed the land of the plaintiff at 267-271 Main Street at $109,720 and the building thereon at $141,300. The plaintiff appealed to the board of tax review, and from the refusal of the board to reduce the assessment it appealed to the Court of Common Pleas. In the course of the trial, the plaintiff withdrew its appeal so far as it related to the valuation of the building. The trial court reduced the assessment on the land to $78,000, and the defendant has taken this appeal.
The property is in the so-called 100 per cent retail business district of the city. On September 1, 1956, it was used for the conduct of the plaintiff's banking business, but its highest and best use, as found by the court and not disputed by the defendant, was for retail business purposes. Although assessments were required by law to be made at the true and actual value of the property (Rev. 1949, 1738), it was the practice of the assessors to fix assessments at 60 per cent of the value as determined by them. This practice, declared illegal in E. Ingraham Co. v. Bristol, 144 Conn. 374, 380, 132 A.2d 563, was validated by the General Assembly at its 1957 session. Public Acts 1957, No. 673, 1; see Rev. 1958, 12-64. In 1955, the assessors undertook the revaluation of real estate which is required during each period of ten years by what is now 12-62 of the 1958 Revision. The valuations fixed by the assessors on the list of 1956 were the result of this revaluation. To aid the assessors in the revaluation, the city engaged the services of the J. M. Cleminshaw Company of Cleveland, Ohio, a nationally recognized appraisal company, and to assist in the task a citizens' advisory committee was formed, consisting of persons familiar with property values in the city. The president of the plaintiff was a member of the committee, as was the expert who testified for the plaintiff. As a result of the revaluation, the basic unit valuation of land in the 100 per cent business area was fixed at $4200 per front foot for a depth of 100 feet. The plaintiff's property has a frontage of approximately 43.5 feet and an average depth of approximately 90 feet, with an easement over land of others for access to the rear of the building. The application of the basic unit valuation to the plaintiff's property, adjusted slightly downward by the use of a depth table to compensate for the 90-foot depth and for the location, which was a short distance northerly from the heart of the 100 per cent business district, resulted in the assessors' setting the true and actual value of the land at $182,866 and the assessment, being 60 per cent of the valuation, at $109,719.60.
At the trial, the only witness testifying for the defendant was an employee of the Cleminshaw Company who had supervised and directed the revaluation. His experience came from employment with the company for a period of fourteen years in the field of reassessment for municipal taxation purposes and from study of the methods of real estate appraisal. The valuation made by him of the plaintiff's property in the course of the revaluation, together with his methods, findings and opinions in fixing that valuation, was accepted and adopted by the board of assessors. In determining the value of the plaintiff's property, he relied on (1) correlation and comparison of two land sales; (2) residual land value capitalization of all property in the central business district; and (3) the advice and recommendation of the citizens' advisory committee. The plaintiff's expert witness fixed the true and actual value of the land at $108,000. He testified that in arriving at this value he used a land residual method. In his opinion the two sales referred to by the defendant's expert were not sales of comparable property and there were no comparable sales available as aids in determining the value of the plaintiff's property.
The trial court found that the two land sales to which the defendant's expert referred were of little use in fixing the value of the plaintiff's property and that there had been no sales of comparable real estate for many years prior to September 1, 1956. It found further that since the criteria of comparable sales were unavailable, the present true and actual value of the property was required to be determined largely by the method employed by both experts — capitalization of stabilized net income. Under this method, the value is determined by capitalizing the future net income from a building — devoted to the highest and best use of the land up on which it is situated — at a capitalization rate which provides for both a return on the investment and a recapture or amortization of the investment over the period of the economic life of the building. The value of the land is determined by the land residual method, which involves capitalization of the income regarded as attributable to the land. This is the amount of the income remaining from the gross rental after proper allowances have been made for vacancies, amortization of the investment in the building, payment of all expenses in connection with its operation, including taxes on the building, and earnings to be derived from ownership of the building. The prevailing rate of return on the investment plus an anticipated tax rate is then used to capitalize the remaining income. This method was used by the plaintiff's expert as well as by the defendant's expert.
The ultimate question presented was the ascertainment of the true and actual value of the plaintiff's property. The parties stipulated in open court that the valuation of, and assessment on, the building was correct, so that there remained for determination only the value of the land: The best test for the determination of value is ordinarily that of market sales. Campbell v. New Haven, 101 Conn. 173, 185, 125 A. 650. Here the court found that there had been no sales of comparable real estate for many years, and this finding was not challenged. In the absence of such sales, other means are required to be employed. Underwood Typewriter Co. v. Hartford, 99 Conn. 329, 337, 122 A. 91. Various methods have been recognized to be proper, no one method being controlling. National Folding Box Co. v. New Haven, 146 Conn. 578, 585, 153 A.2d 420; Sibley v. Middlefield, 143 Conn. 100, 107, 120 A.2d 77. The experts who testified for each of the parties agreed that the best method to be employed here was the land residual method. In that method, as in the case of other theoretical methods of establishing value, a number of factors and elements are involved and require consideration, so that mathematical accuracy is not necessarily achieved by their use. Lomas Nettleton Co. v. Waterbury, 122 Conn. 228, 232, 188 A. 433. In situations where such mathematical calculations must be relied upon for the determination of value, much caution is required, because those calculations, although made in the best of faith, can lead to widely divergent results. They must, therefore, be closely scrutinized. In re Erie Railroad System, 19 N.J. 110, 135, 115 A.2d 89. The methods do not establish value in themselves, but are merely aids in the ascertainment thereof.
The assessment of property for taxation is an administrative proceeding. The board of assessors and the board of tax review are administrative boards, and in performing their duties in valuing property and in correcting and equalizing assessments, they act largely upon the knowledge of their members as to valuation and the taxable property of taxpayers. Bugbee v. Putnam, 90 Conn. 154, 158, 96 A. 955; Ives v. Goshen, 65 Conn. 456, 459, 32 A. 932. In considering the results arrived at by them, 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. Bridgeport Brass Co. v. Drew, 102 Conn. 206, 212, 128 A. 413; Ford v. H. W. Dubiskie Co., 105 Conn. 572, 580, 136 A. 560. Valuations for taxation purposes are first fixed by the assessors and an appeal lies to the board of tax review. The law contemplates that a wide discretion is to be accorded to assessors, and unless their action is discriminatory or so unreasonable that property is substantially overvalued and thus injustice and illegality result, their opinion and judgment should control in the determination of value for taxation purposes. Sears, Roebuck Co. v. Presque Isle, 150 Me. 181, 189, 107 A.2d 475; German-American Lumber Co. v. Barbee, 59 Fla. 493, 498, 52 So. 292; Washington County National Bank v. Washington County, 176 Va. 216, 222, 10 S.E.2d 515; Haubrich v. Johnson, 242 Iowa 1236, 1246, 50 N.W.2d 19.
It has been noted repeatedly that the valuation of property for assessment purposes has been a vexing and much litigated problem. Sibley v. Middlefield, 143 Conn. 100, 105, 120 A.2d 77. A taxpayer who thinks he is aggrieved has an appeal to the courts, where the matter is tried de novo. On such an appeal, the contesting taxpayer must establish that the assessment is unjust and will result in his being required, in effect, to pay an illegal tax. Sibley v. Middlefield, supra; Ives v. Goshen, supra. The burden of proof is on him. Thaw v. Fairfield, 132 Conn. 173, 179, 43 A.2d 65. Here, since the value of the plaintiff's land was required to be determined largely by use of a mathematical process, the burden which rested on the plaintiff Was to establish that the factors employed in that process were improper and resulted in an unjust and illegal assessment.
The trial court found that the vital factors in the use of the land residual method are the capitalization rate and the method of recapturing the investment in the building, or amortization. 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. 1 Bonbright, Valuation of Property, p. 259. The assessors, upon the recommendation of the defendant's expert, used a capitalization rate of 5.5 per cent. This rate was based on a number of factors, including the assumption that a mortgage could be obtained on the property for 80 per cent of its value at 4.5 per cent interest. The court found that a capitalization rate of 5.5 per cent could not be supported on September 1, 1956, since the only credible evidence was that no lending institution would have made a mortgage of 80 per cent at 4.5 per cent interest on that date and that under conditions then prevailing only a mortgage to the extent of 60 per cent, at 5.5 per cent interest, would have been available. This was a crucial finding, and it was challenged by the defendant. The only evidence to which the plaintiff has pointed to support this finding is found in the testimony of the plaintiff's president. To a question by the court whether, from the experience of the witness and his knowledge of conditions existing on September 1, 1956, an investor could obtain a 4.5 per cent mortgage for 80 per cent of the value, the witness replied: "Well definitely speaking for our own institution, that would be impossible. I cannot speak for others of course." He was then asked to testify from his general experience as a banker, and he replied, "I do not believe you could get a first mortgage at [4.5 per cent on 80] per cent of the value of the property. I think it would have to be a combination of first and second mortgages to get as high as 80 per cent." It is obvious that this testimony does not support a finding that no lending institution would have made a mortgage of the nature in question. On the other hand, in contrast to this testimony, there was testimony by the defendant's expert that insurance companies would be interested in making such loans, and he referred to a loan made to a shopping center in Hamden at 4.5 per cent interest.
Moreover, if the availability of such mortgage loans was an essential factor in fixing the capitalization rate, the test was not merely whether such loans were available on September 1, 1956. That date fixed the date as of which assessments were required to be made. But the elements and factors required to be considered in determining value were such as existed during a reasonable period prior to the assessment date. Somers v. Meriden, 119 Conn. 5, 12, 174 A. 184. "`Values of real estate and fixtures thereon are more or less constant over a period of years. . . . While the assessment is to be made as of a certain date, the value of the property is established over a period of years.'" Id., 10, quoting from Central Realty Co. v. Board of Review, 110 W. Va. 437, 440, 441, 158 S.E. 537. Tax assessors are required to recognize and act on the principle that the true value of a fixed asset such as real estate is fairly constant and must be gauged, not by conditions temporary and extraordinary, but by those prevailing over a period of time, and the assessors, in listing values of property for taxation, may, to a certain extent, disregard the excesses of a boom as well as the despair of a depression. Alfred J. Sweet, Inc. v. Auburn, 134 Me. 28, 32, 180 A. 803. It is well known that interest rates fluctuate and lending policies vary from time to time in accordance with general economic conditions. The burden that rested on the plaintiff was to establish that, over a reasonable period of time, no mortgage of the type assumed by the assessors to be available was in fact obtainable. No such evidence was presented. The finding by the court that a mortgage of the type in question was not available is without support and may not stand. In the absence of this finding, the plaintiff's attack upon the propriety of the use by the assessors of a 5.5 per cent capitalization rate must fail. There was error in the court's conclusion that the assessors were not justified in using a 5.5 per cent capitalization rate in their calculations.
The other vital factor in the land residual method is a proper method of recapture of the investment in the building. Testimony was presented as to three recognized methods. One of these methods, known as the straight line method, was employed by the plaintiff's expert. Another method, known as the Hoskold method, was employed by the defendant's expert. The straight line method provides for recapture of the same amount every year over the economic life of the building. Under the Hoskold method, a fixed amount is transferred each year into a sinking fund and compounded at a given rate of interest during the economic life of the building. At the end of that period, the full amount of the investment in the building becomes available. In the instant case the defendant's expert fixed the interest rate at 3.5 per cent. "[N]o single formula for estimating depreciation is free from serious error, save in a few selected instances." 1 Bonbright, Valuation of Property, p. 203. The plaintiff's expert used a sixty-year amortization basis with the straight line method, while the defendant's expert used a forty-year amortization basis with the Hoskold method. Amortization by the straight line method, assuming an economic life of fifty years for the building, is at the rate of 2 per cent a year, while by the Hoskold method, as computed by the defendant's expert and assuming an economic life of forty years, it is at the rate of 0.76 per cent a year. This difference in the rate of amortization would alone result in a difference in the estimated value of the plaintiff's land of $20,000.
The court found that the prevailing method for amortizing the capital investment in real estate such as the plaintiff's in New Britain on September 1, 1956, was the straight line method as distinguished from the sinking fund method. The only evidence which may, in any way, be said to support this finding is in the testimony of the plaintiff's expert, who said that his own appraisal of the property was based upon the straight line method of recapturing capital. The fact that the witness chose to use the straight line method in his own calculations and appraisal falls far short of supporting a finding that this method, or any other single method for recapturing the capital investment in real estate, prevailed in New Britain. Here, too, the finding was crucial. It was the basis for the court's conclusion that the use of the Hoskold, or sinking fund, method by the assessors was improper. The testimony of the defendant's expert was to the effect that the Hoskold method was the correct one to employ and that it met all the tests referred to by the American Institute of Real Estate Appraisers for testing the validity of a theory of appraisal. The court, of course, was not required to give credence to the testimony of the defendant's expert: But its failure to accept his testimony would furnish no support for a contrary finding that the straight line method was the prevailing method employed in New Britain and therefore was the method required to be employed here. State v. Poplowski, 104 Conn. 493, 495, 133 A. 671. The finding lacks support in the evidence, and on the record the court erred in employing the straight line method of amortization in arriving at its determination of value.
The court, in an exhaustive memorandum of decision as well as in other portions of the record, appears to have been aware of the problems presented by reason of the necessity, in this case, of relying on mathematical calculations to such a large degree in the ascertainment of the value of the plaintiff's land. The determination of the value for taxation purposes of the property under review, in an appeal such as this, is ordinarily left as a question of fact for the trier. National Folding Box Co. v. New Haven, 146 Conn. 578, 588, 153 A.2d 420; Connecticut Savings Bank v. New Haven, 131 Conn. 575, 584, 41 A.2d 765. Such a determination by the trial court presupposes, however, that the evidence has established a basis for the exercise of the broad discretionary power which the statute authorizes for the correction of an illegal assessment. Sibley v. Middlefield, supra, 105. On the record here, the findings essential to establish that the methods employed by the assessors resulted in an excessive and illegal assessment lack support in the evidence, and the court erred in concluding that it was empowered to grant the relief sought.