179 N.J. Super. 373 (1980)
432 A.2d 127
ABE GAYNES T/A MARINA GARDENS, PETITIONER-RESPONDENT,
v.
TOWNSHIP OF EDISON, RESPONDENT-APPELLANT.
Superior Court of New Jersey, Appellate Division.
Argued February 11, 1980.
Findings and Conclusions June 25, 1980.
Reargued September 29, 1980.
Decided October 8, 1980.
Before Judges SEIDMAN and LANE.
Peter A. DeSarno argued the cause for appellant.
Edward G. Rosenblum argued the cause for respondent (Rosenblum & Rosenblum, attorneys; Leo Rosenblum of counsel).
Argued February 11, 1980 Remanded to Tax Court for further proceedings.
Findings and Conclusions of Tax Court June 25, 1980.
PER CURIAM.
The Township of Edison appeals from three judgments entered in the Division of Tax Appeals reducing the assessments for the years 1974, 1975 and 1976 on lands and improvements owned by the petitioner, Abe Gaynes, t/a Marina Gardens. The original assessment on the land for each year was $557,800. The improvements were assessed at $2,146,200 for 1974 and 1975, and $2,727,900 for 1976. Appeals to the county board of taxation were dismissed. On further appeal to the Division of Tax Appeals the judge there determined that the fair market value of the lands and improvements for each year was $2,633,242. He found, further, that petitioner had proved its claim of discrimination with respect to the assessment and, using a common level factor of 75%, reduced the assessment to $1,974,932, allocating it $557,800 to the land and $1,417,132 to the improvements.
On this appeal, the township contended that the taxpayer's proofs "failed to satisfy the criteria set forth in Tri-Terminal Corporation v. The Borough of Edgewater [68 N.J. 405 (1975)]," and, therefore, the granting of discrimination relief was improper; that the best indication of market value was the sales of the subject property, and that the court below erred in determining the common level.
The judge of tax appeals heard conflicting evidence on the fair market value of the property, which consists of a 10.954-acre tract on which is constructed a 201-unit garden apartment complex. The taxpayer's expert valued the property at $2,260,000, arriving at the figure by capitalizing the net income, computed by him to be $303,111, using a capitalization rate comprised of 8 1/2% interest, 3% recapture and an average effective tax rate of $2.95. An expert appraiser testifying for the municipality also relied chiefly on the income approach. His net income figure was $339,768, which he capitalized with a rate consisting of 8% interest, 2% recapture and an average tax rate of $3.51. His resulting value was $2,600,000. The tax judge rejected the sales prices of the subject property as insufficient market data "due to the obvious financial manipulations involved in them." On the issue of value, he also chose the income approach. He took the gross income figure of $520,000 used by both experts and, accepting as "better proof" average expenses of 34.66%, capitalized the net income of $339,768. Using a rate consisting of 8% interest, 2 1/2% recapture and an effective tax rate average of $2.97 for the three years in question, he determined the fair market value of the land and improvements to be $2,633,242.
Although petitioner is content with the finding of fair market value, acknowledging in its brief that it is supported by substantial evidence in the record, the township maintains that the market data approach was the fairest way to appraise the property, even though its own expert relied on the income approach. The proofs disclosed that petitioner sold the property in 1973 for $2,900,000 with three mortgages covered by a wraparound fourth mortgage for $2,500,000; that the purchaser resold the property a year later for $3,000,000, with $100,000 in cash being paid plus a fifth mortgage, and that when the new owner defaulted in 1975, petitioner became the receiver of the property and the prior purchases were wiped out. In the circumstances the judge of tax appeals was amply justified in refusing to consider the sales as satisfactory evidence of the fair market value of the subject property. The findings and conclusions of the judge with respect to the fair market value of the property could reasonably have been reached on sufficient credible evidence present in the record, considering the proofs as a whole and giving due regard to the ability of the factfinder to judge of the credibility of the witnesses. Passaic v. Botany Mills, 72 N.J. Super. 449, 454 (App.Div. 1962).
On the issue of discrimination, the judge of tax appeals found, with support in the record, that while the subject property had been taxed at 100% of true value "or very close to it," the common level of taxation in the township was "well below it." "Common level" has been defined as "a ratio to or percentage of full true value at which property generally [is] assessed in the municipality." In re Appeal of Kents, 34 N.J. 21, 26 (1961).
At the prior argument of this appeal a dispute arose regarding the method used to arrive at the common level of 75%. We requested supplemental letter briefs on the issue. The taxpayer argued that in determining the capitalization rate the judge of tax appeals used the average tax rate for the three applicable years multiplied "by the average weighted rates"[1] and, after reaching his judgment of value based on this approach, he then multiplied the true value by 75% to eliminate discrimination. The complaint is that the 75% represented the approximate unweighted average ratio and, to be consistent, the judge of tax appeals should have used the same ratio for both purposes. Had he done so, according to the taxpayer, the ultimate assessment would have been $2,035,196 instead of $1,974,932. Despite this generous concession by the taxpayer, the township contended that the weighted ratio should have been applied to the true value as found, resulting in an assessed valuation of $2,229,303.
Since these arguments were not advanced until oral argument, we chose to defer to the expertise of the Tax Court and, to that end, in an unreported opinion, we remanded the matter to that Court for such further proceedings "as may be deemed necessary to a resolution of the conflicting arguments presented by counsel," and for the filing thereafter in this court the findings and conclusions.
Such proceedings were had and in due course Tax Court Judge Andrew filed his findings and conclusions. Judge Andrew found that the common level in the township for the three tax years involved "was at the unweighted ratio averaged ... at 75.72%." The Tax Court judge further determined that "the unweighted, unclassified arithmetic ratio should be applied to eliminate inequality in assessment in this case if such inequality is found to exist." He concluded that in the income approach to value the effective tax rate must reflect the unweighted and not the Director's weighted ratio "to insure consistency." Accordingly, he recomputed the capitalization figures used by the judge of tax appeals and determined that the assessed value should be $2,035,200, allocated $422,400 to land and $1,612,800 to improvements. We agree with the result reached substantially for the reasons expressed by Judge Andrew in his findings and conclusions.
The judgment of the Division of Tax Appeals is modified accordingly and, as modified, is affirmed.
FINDINGS AND CONCLUSIONS
ANDREW, J.T.C.
This matter raises for the first time since In re Appeal of Kents, 2124 Atlantic Ave., Inc., 34 N.J. 21, 32 (1961), the issue of whether the unweighted average sales-assessment ratio of the Director of the Division of Taxation is better suited than the Director's weighted average sales-assessment ratio for the purpose of providing relief to a taxpayer in a local property tax case involving inequality of assessment.[1]
Plaintiff challenged its local property tax assessment on the ground that it exceeded fair market value and also contended that the assessment should be reduced to the level of assessments generally prevailing in the municipality for the tax years of 1974, 1975 and 1976. After a hearing before the Division of Tax Appeals, the trial judge made a value determination and also found that discrimination in assessment existed, therefore entitling plaintiff to Kents type relief. Ibid. Inasmuch as the property involved was an income-producing garden apartment complex, the Division judge relied upon the income approach to value as being the most persuasive tool for deriving fair market value.
In his utilization of a reconstructed income approach he used the Director's weighted sales-assessment ratio in arriving at an effective tax rate, pursuant to the requirement of New Brunswick v. Tax Appeals Division, 39 N.J. 537, 547 (1963). However, after finding true value he then applied the Director's unweighted, unclassified sales-assessment ratio in order to eliminate the inequality in assessment. The taxing district appealed the judgments entered in the Division of Tax Appeals, and at oral argument before the Appellate Division of the Superior Court the issue of the trial judge's use of two different ratios was posed for the first time. The Appellate Division referred to the language of our Supreme Court in In re Appeal of Kents, supra, wherein it was noted:
We should add that we are aware of some opinion that the unweighted average ratio is better suited than the weighted average ratio for the purposes of an individual taxpayer's appeal. We leave the initial determination of the subsidiary question to the expertise of the State Division. [34 N.J. at 32]
As a result of the exceptions taken by both parties to the use of both the weighted and unweighted ratios by the trial court, the Appellate Division remanded the matter to the Tax Court for findings and conclusions in order to resolve the dispute as to the proper ratio to be applied.
A counsel conference was held by this court and it was stipulated that the court would base its findings and determinations on the record established before the trial judge supplemented by written briefs and oral argument.
The record reveals that the taxpayer's expert was of the opinion that the unweighted, unclassified ratio reflected the common level in Edison during the tax years in question. He stated that if the term "common level" meant that "single ratio to true value at which the great bulk of the ratables is assessed" or that "ratio to or percentage of full true value at which property generally was assessed in the municipality," then it can be concluded that Edison had a common level of assessment which was the level of the unweighted, unclassified sales-assessment ratios averaged for the three tax years. The trial judge agreed with this conclusion and found that the common level was the unweighted ratios averaged for the three tax years.
Even though his opinion advocated the use of the unweighted, unclassified ratio, the taxpayer's expert developed different effective tax rates based on the weighted ratio and on the unweighted ratio and ascribed differing values to the subject property as a result. He did this because the Division of Tax Appeals and appellate tribunals in the past have relied upon the weighted ratio. He could not indicate one reported decision that utilized the unweighted ratio. He repeated this procedure in his determination of a proper assessment, i.e., he applied the weighted ratio and unweighted ratio to his true values to arrive at two different assessment conclusions. The municipality did not present testimony at the hearing relative to either ratio. It relied upon the accepted use of the weighted ratio by the courts in New Jersey since In re Appeal of Kents, as being dispositive of the issue.
Our Supreme Court, subsequent to In re Appeal of Kents, mentioned the subject of weighted and unweighted ratios in a footnote to an opinion in Tri-Terminal Corp. v. Edgewater, 68 N.J. 405, 413 (1975). Therein Judge Conford, speaking for the court, stated that the unweighted ratio was preferable to the Director's weighted ratio, but he did not go into an explanation for its use.
Before an analysis is made of the two ratios, an explanation of the methodology of the Director of the Division of Taxation in arriving at the specific ratios will assist in understanding the issue presented.
The Local Property Tax Bureau of the Division of Taxation reviews all sales reported to it by assessors and county boards of taxation and determines whether each sale is usable or nonusable based on specific valuation criteria. By this process the Bureau seeks to verify that a sale is a bona fide, arm's-length transaction. This is done in order to eliminate any transactions which were influenced by nonmarket oriented factors such as a transfer from a wife to a husband.
The usable sales are then divided into four categories. Class 1 is vacant land, class 2 is one- to four-family residential property, class 3 is farmland and class 4 includes commercial, industrial and residential designed for five or more families. A ratio is determined for each of the four classes by dividing the total assessed value of all properties sold during the sampling period by the total of the sales prices of the properties. The true value of the real property in each class is then determined by dividing the aggregate assessed value of each class by the class ratio.
The weighted classified ratio for a municipality is then found by adding the assessed valuations for each class and dividing that sum by the total of the true values for each class. The sampling period for the Director's weighted ratio is July 1 of the pretax year to June 30 of the tax year. The ratio derived from the sales during this time is then averaged with prior years on a regression basis. The latest sales period is averaged with the previous year's sales period. The current year is given a weight of 50% and the previous year is given a weight of 25%. The year previous to the second year is also considered, giving that year a weight of 12 1/2% and each prior year is considered with half the weight of the succeeding year until there are no significant percentages remaining.
The weighted ratio also known as the Director's ratio, thus determined, is that which is used for the distribution of state school aid pursuant to N.J.S.A. 54:1-35.1. As Chief Justice Weintraub indicated in In re Appeal of Kents, supra:
The State Director's average ratio ... [is] primarily intended to meet the problem of intermunicipal inequality.
[It is] designed to establish the total true value of the aggregates of real property within each municipality ... for the purpose of fixing a basis for the distribution to municipalities of State aid for education.... [at 26]
The Chief Justice warned that the weighted ratio is not that handy tool which would enable an assessor to achieve a revaluation or reassessment while sitting at his desk. Id. at 27. Judge Conford, in his dissenting opinion in Piscataway Assoc, Inc. v. Piscataway Tp., 73 N.J. 546 (1977) repeated the admonition by stating:
These sales studies are conducted by the Director solely for the purposes of aiding distribution of state aid to school districts and as guides for establishment of county equalization tables. Tp. of Willingboro v. Burlington Cty. Bd. of Tax, 62 N.J. 203, 209 (1973). They have nothing to do with the setting of individual property tax assessments. [73 N.J., at 557]
After the same screening process which is used in the formulation of the weighted ratio, the unweighted, unclassified ratio is determined by dividing each assessment of a property sold during the sampling period by the sales price to produce an assessment-sale ratio for each sale. All of the assessment-sales ratios are then added to produce a total which is divided by the total number of sales. As can be seen, the ratio is unweighted because the size or amount of each assessment is unimportant. Each individual ratio has the weight of 1. The sales sampling period for this ratio covers July 1 of the pretax year to June 30 of the tax year.
The Local Property Tax Bureau of the Division of Taxation has provided some guidance to assessors as to when a revaluation should be sought within a taxing district in order to achieve equality of assessment within the district. Handbook for New Jersey Assessors (rev. ed. 1965), at 201 (by Bureau of Government Research, Rutgers, the State University). The Bureau points to the sales-assessment ratio studies and the coefficient of deviation which is a statistical tool that assists in analyzing the results of the sales studies. The coefficient measures the deviation from the unweighted, unclassified ratio within a taxing district. If a coefficient is low, there is an indication that most of the individual assessment-sales ratios fall within a narrow range and the better the assessing practice in a municipality hence substantial equality may obtain. However, if the coefficient is high, there may be need for revaluation. The Handbook states (at 203) that a coefficient of deviation of less than 20% is considered good, meaning that a revaluation is not required. It is also noted in the Handbook that all sales should be given equal weight regardless of the classification of the property. Ibid. Since the coefficient is measured from the unweighted ratio and not the weighted ratio it also provides for equal weight regardless of the size of the assessment.
The ratio data does not provide mathematical perfection in levying assessments, but such precision is not attainable in the valuation of property for assessment purposes nor is it required. Baldwin Constr. Co. v. Essex Cty. Bd. of Tax., 16 N.J. 329, 342 (1954). The sales ratio studies are available to assist in providing relief against inequality when an aggrieved taxpayer has an assessment substantially above the common level of assessments within a taxing district. The common level has been defined as "a ratio to or percentage of full true value at which property generally [is] assessed in the municipality," or "a single ratio to true value at which the great bulk of the ratables is assessed...." In re Appeal of Kents, supra, 34 N.J. at 26.
In our search for that ratio which is a better indicator of the common level of assessments, we note that the taxpayer has presented an example which illustrates the distortion which could be inherent in the weighted ratio. He asks the court to assume that a taxing district has a total of 1,001 parcels of real property, 1,000 of which are identical single-family homes, each having a market value of $20,000 and an assessed value of $10,000. This would result in a ratio of assessed to true value for each of 50%. The 1,001st parcel is a large industrial complex that has a market value of $20,000,000 (equal to the value of the 1,000 homes combined). The assumption is that this industrial property is assessed at 100% of its market value. The last assumption is that all parcels are sold during a one-year sampling period at their putative market values.
The average weighted ratio which would be produced by this data in accordance with the method heretofore described for determining the Director's ratio would produce a sales-assessment ratio of 75% for the one-year sampling period. The weight of the one industrial assessment would provide a deviation of 50% from the ratios of the single-family homes. (75% as to 50%).[2] Utilizing the unweighted procedure would produce a ratio of 50.05% overall since the industrial assessment-sale ratio would receive no more weight than the ratio for each of the individual single-family homes.
The taxpayer concludes by stating that this hypothetical clearly establishes that the unweighted ratio better reflects the common level because it is that percentage of full true value at which most of the properties in the taxing district are assessed. The distortion extant in the hypothetical was produced by one sale. This could not happen with the unweighted ratio because each individual assessment-sales ratio has the same weight, a weight of 1.
Assuming that the number of sales for a given sampling period is adequate, concededly the unweighted ratio gives a more equitable indication of the level of assessment generally prevailing in a municipality during that particular period. The record indicates that there was a sizeable number of sales during the three-year period in question.[3] The township did not contend that there was a weakness in the sales-assessment ratios because of a paucity of sales during any one year or totally.
I find based on the record produced at the hearing before the Division judge and the briefs and oral argument that a common level did exist within Edison during the tax years of 1974, 1975 and 1976. However, it is clear that this common level was not 100%. The ratio clusters indicate a level within the range of 60% to 79%, supporting the opinion of the expert for the taxpayer that the common level was at the unweighted ratio averaged for the three tax years at 75.72%. This conclusion is further supported by the coefficients of deviation for each tax year. This statistical barometer indicates that the assessing practice was good in that the general or overall coefficients were less than 20% as follows:
1974 12.46
1975 11.68
1976 11.47
As previously stated, the coefficients measure the assessment practice and indicate the existence or nonexistence of a common level. The coefficients for Edison indicate a common level at the level of the unweighted ratio, which is the norm against which the deviation in individual assessment-sales ratios is measured. There is little question that the sales sampling (1533 sales for three tax years) was adequate to demonstrate that level at which the great bulk of ratables is assessed. The record is devoid of any evidence by the township that would contravene the conclusion of the taxpayer's expert that the unweighted ratio constituted the common level of assessment within the taxing district.
It should be noted at this juncture that the present factual pattern differs somewhat from In re Appeal of Kents, supra. In Kents it was clear that there was a total absence of any semblance of a common level of assessment. The court considered it appropriate to provide relief because the weighted ratio was deemed sufficient evidence of the level to which a reduction should be granted. Id. at 30. Here, there is an allegation and proof of a common level. The informative language of our Supreme Court in In re Appeal of Kents is apposite:
If the sales data used to find the ratio in fact revealed some percentage of true value about which the bulk of individual assessments tended to cluster, one might accept that percentage as the common level of assessment. [at 27]
However, it is the opinion of this court that the unweighted ratio should be utilized considering the facts in this case, regardless of whether a common level is found to exist or a determination is made that a common level does not exist in Edison.
In response to the contention of the taxpayer that the unweighted ratio should be utilized to eliminate inequality in assessment, the township asserts that there is no reported case that mandates which ratio should be applied and that the practice of the Division of Tax Appeals was to accept the weighted ratio. I can only assume that no one sought to litigate the issue and therefore it has not been squarely presented as it has in the present matter.
Edison further contends that the Legislature has provided the guidepost to follow in its amendment of N.J.S.A. 54:1-35a wherein the weighted ratio was substituted for the unweighted ratio in the application of N.J.S.A. 54:2-40.4 (commonly known as chapter 123, L. 1973, c. 123). Defendant submits that therefore the standard to be applied in discrimination in assessment cases is the weighted ratio. To apply any other ratio would create different results depending solely upon the year of the proceeding.
The standard as promulgated does not apply to this matter for a number of reasons. Initially N.J.S.A. 54:1-35a and N.J.S.A. 54:2-40.4 were not effective for any of the tax years in question. The cited statutes became applicable with the tax year of 1978. The original provision prescribed the use of the unweighted, unclassified sales-assessment ratio. This, as previously stated, was then amended to provide that the weighted ratio would be applied to the 1979 tax year and thereafter. L. 1979, c. 51. There is no requirement in the law that this yardstick be applied to pre-enactment years. On the contrary, it specifies the tax years to which the stated ratios are to be applied.
The argument that use of the unweighted ratio would create different results in different years because of N.J.S.A. 54:1-35a and N.J.S.A. 54:2-40.4 is without merit. If one were to follow this contention to its logical conclusion, the court would apply the weighted ratio in 1976, the unweighted ratio in 1978 and the weighted ratio in 1979.
This court, however, has on its own motion raised the question of stability of assessments. Our appellate tribunals have noted that as a practical matter there must be a certain measure of stability in assessments. New Brunswick v. Tax Appeals Director, 39 N.J. 537, 541 (1963); Feder v. Passaic, 105 N.J. Super. 157, 162 (App.Div. 1969); Samuel Hird & Sons, Inc. v. Garfield, 87 N.J. Super. 65, 76 (App.Div. 1965).
The Statement of the Senate County and Municipal Government Committee annexed to Assembly Bill 1492 (L. 1979, c. 51) provided that one of the reasons for the amendment by the Legislature from the unweighted ratio to the weighted ratio was that the latter ratio was averaged over the latest two years. This averaging method would avoid abrupt changes in ratio from year to year and would also avoid an undue influence of a dearth of sales in any single year.
These considerations might have merit were it not for the fact that here we are dealing with three tax years and an indicated sufficiency of sales data. Pursuant to Feder v. Passaic, supra, the average of the three unweighted, unclassified ratios should be used. This would provide stability and a more than adequate sales sampling base.
The only other concern which this court has raised, on its own motion, is the question of whether the ratios which should be used are those as of the assessment dates, i.e., October 1 of the pretax year, or the ratios which are promulgated as of the tax year in question. The Appellate Division used the ratios of the State Director promulgated during the tax years in question for those tax years. Ibid. The merit in this approach can be seen in the fact that the latest sales samplings are being used to reflect the current common level. The level for 1974 will be indicated by sales that occurred during the period of July 1, 1973 to June 30, 1974, rather than the previous year's sales if we were to adopt the ratio promulgated on the assessment date of October 1, 1973 to be applied to the 1974 tax year. It would seem that the latest sales would be the better determinant of common level during a particular tax year.
Based on all of the foregoing observations, it is the conclusion of this court that the unweighted, unclassified arithmetic ratio should be applied to eliminate inequality in assessment in this case if such inequality is found to exist.
If the unweighted ratio is to be applied to provide relief from discrimination, then in the income approach to value the effective tax rate must reflect the unweighted ratio and not the Director's weighted ratio to insure consistency.
Accepting the findings of the trial judge indicating that an economic analysis should be used and his reconstructed income statement save for the effective tax rate, the value produced thereby would be as follows:
Net Operating Income $ 339,768
Deduct Income to Land
($557,000 × 10.66%)
(8% return + 2.66% tax) 59,461
(average tax rate of $3.51 × 75.72%)
_________
Net Income to Improvements $ 280,307
$280,307 capitalized at 13.16%
(8% return, 2.5% recapture and 2.66% tax) 2,129,992
Add Value of Land 557,800
__________
Value for 1974, 1975 and 1976 (rounded) $2,687,800
Application of the common level of 75.72%, which is the average of the unweighted sales-assessment ratios for the three tax years, to the value will produce rounded assessed values as follows:
Land $ 422,400
Improvements 1,612,800
__________
Total $2,035,200
NOTES
[1] There are two methods of determining the average ratio of assessed to true value of real property. One is the "unweighted," unclassified, statistical approach, whereby the average ratio is determined by adding together all sales ratios to true value during a two-year study period without regard to classification of the properties involved, and dividing the total by the number of sales involved in the study. Perth Amboy v. Middlesex Cty. Bd. of Tax., 91 N.J. Super. 305 (App.Div. 1966). The "weighted," classified method, used by the Director of Division of Taxation each year in preparing the tables of equalized values for use in allocating state aid to local schools, takes into account four different classes of real estate, thus giving more weight, for example, to a single large sale in one class.
[1] In Perth Amboy v. Middlesex Cty. Bd. of Tax., 91 N.J. Super. 305 (App.Div. 1966), the court had before it the question of the use of the unweighted ratio as opposed to the weighted ratio with regard to the county equalization table. The court held that the use of the unweighted ratio for determining apportionment of the inter-municipal cost of the county government was not unreasonable and therefore affirmed its use in Middlesex County for equalization purposes.
[2] Although this distortion in the current year could be reduced in effect by the averaging of ratios with prior years, as is done by the Director, it would have to be assumed that there was no distortion in the prior years.
[3] The sales ratio study introduced into evidence by plaintiff revealed 598 usable sales in 1974, 475 usable sales in 1975 and 460 usable sales in 1976 a total of 1533 usable sales, which is a sufficient sampling.