PARKVIEW VILLAGE ASSOCIATES, PETITIONER-APPELLANT,
v.
BOROUGH OF COLLINGSWOOD, A MUNICIPAL CORPORATION OF THE STATE OF NEW JERSEY, AND DIVISION OF TAX APPEALS OF THE DEPARTMENT OF THE TREASURY OF THE STATE OF NEW JERSEY, RESPONDENTS.
The Supreme Court of New Jersey.
Argued November 6, 1972. Decided December 18, 1972.*23 Mr. William J. Walsh, argued the cause for appellant (Mr. William J. Walsh, on the brief; Mr. Frank V. Walsh, Jr., attorney).
Mr. Eivind H. Barth, Jr. argued the cause for respondent Borough of Collingswood.
The opinion of the Court was delivered by CONFORD, P.J.A.D., Temporarily Assigned.
This is a tax appeal involving an assessment for the year 1968 (assessing date October 1, 1967) of petitioner's large apartment house complex in Collingswood, Camden County. The property consists of four buildings, two of nine stories and two of ten, containing 1,036 apartment units, commercial space and a garage. The assessments (based upon a 50% common level factor), which are separate for each of the four building lots, aggregate $177,500 for land and $4,937,650 for improvements, or a total of $5,115,150 (true value, $10,230,300).
The Camden County Board of Taxation affirmed the assessments while the State Division of Tax Appeals reduced them on appeal to an aggregate of $5,058,500 (true value, $10,117,000), or a reduction of $56,650, all on improvements. The Appellate Division affirmed the judgment. On application of the taxpayer this Court granted certification. 60 N.J. 467 (1972).
Determination of this appeal will not require extended consideration of all the factors relied upon by the two experts for respondent and the single expert for petitioner, in arriving at their valuations given in testimony at the hearing, or of those used by the Division judge in his determination. Although each of the experts submitted alternative valuation methods to support their conclusions, all agreed, and the Division concurred, that the income approach should be of preponderant influence in this case, and we are in accord. The core issue is the justification for the estimate by respondent's witnesses of the fair rental value ("economic" rental *24 income) of the property by adding 20% to the contract rental scale as of the assessing date, and for that of the Division in applying a 15% appreciation factor thereto, in arriving at their respective determinations of capitalization value. Although there were variations as among the experts and the Division judge as to the other significant factors employed in capitalizing net income, and petitioner complains of the figures used by the Division in some of these respects, the controlling issue presented relates to the matter of economic rental value of the property.
The discussion should be prefaced by the precautionary note sounded by the Chief Justice in the seminal case in this field New Brunswick v. State of N.J. Div. of Tax Appeals, 39 N.J. 537, 544 (1963) as to "the care with which the income method must be used" and "that one should hesitate to accept its answer without checking against all available data." These observations are peculiarly apposite in this case.
Bate, petitioner's only expert witness, arrived at a total valuation for the property of $8,341,400 through the capitalized income approach. For this purpose he took the contract rental scale as of February 1, 1968 (apparently the taxpayer's fiscal accounting date) and projected that to a full occupancy basis for a total gross apartment income of $1,521,348. (There was also minor income from commercial facilities.) He deemed that figure appropriate because "more in line with economic rentals" than any 1967 scale. In arriving at net rental income one of his deductions was a vacancy factor of 3%, although actual vacancy was somewhat less.
Bate was of the view that there were no other comparable high rise apartments against which to check Parkview rental values. In reference to the Cherry Hill apartments resorted to by the opposing experts for this purpose (see infra), he thought they were "newer and much more modern and more expensive in category with a different type of tenant."
Olasin, respondent's first expert witness, testified that these buildings were completed in 1950 and had a present *25 value on the capitalized income approach of $10,280,000. He thought the only apartment house "anywhere near comparable" to the subject property was a two-building complex in Cherry Hill completed in 1957 on Route 38 across the highway from the massive Cherry Hill Shopping Mall. These apartments are twelve stories high, contain 430 dwelling units and have universal air-conditioning and a swimming pool. The Parkview apartments have neither, but are adjacent to two public parks. Moreover, Olasin conceded that Cherry Hill was nationally known and considered a more prestigious address for an apartment in the Camden area than Collingswood. But there was "terrific turnover" of tenancies in the Cherry Hill community as compared with the "low turnover" characteristic of the Collingswood area. The compared complexes are four to five miles apart, but Olasin deemed Parkview to have better public transportation to nearby Camden and Philadelphia.
Olasin compared the projects from a rental value viewpoint, giving Cherry Hill a 10% preferential factor for prestige and status and 5% for amenities. He gave Parkview a similar factor of 10% for location and accessibility, but rated both projects equal for services. He thus concluded that Cherry Hill had a net 5% higher fair rental value than Parkview. He then purported to compute the square footage of rentable apartment space in each complex and to arrive at an annual square foot rental unit value for each, showing $2.31 for one of the Cherry Hill buildings and $2.25 for the other. Discounting these by 5% yielded an adjusted average Cherry Hill unit value of $2.14 which Olasin concluded was what Parkview should be economically renting for. Since his computed Parkview figure was only $1.79 per square foot, the witness concluded that Parkview's "economic rent" was 20% more than its contracted rental scale and that the latter should be increased by the stated percentage for purposes of capitalizing net income. Although Olasin conceded that rents in both apartments rise with the *26 height of the floor of the apartment, he made no adjustment for the difference in heights of the complexes under comparison.
In his deductions for arriving at net income Olasin applied a vacancy factor of 5%.
As a subsidiary basis for his conclusion as to Parkview's economic rental value Olasin noted the low vacancy experience of Parkview (conceded to average 1 1/2%) and said that fact showed the rents were too low. Cross-examined as to why the high vacancy rate (10%) at the Cherry Hill apartments should not with equal or greater logic have indicated that the rentals charged there were too high, Olasin stated, without furnishing particulars, that management at Cherry Hill had been poor.[1]
Todd, respondent's other realty expert, testified to a capitalized value of $10,200,000. He subscribed fully to Olasin's square foot unit comparison of rental values of Cherry Hill and Parkview. Primarily on that basis he also added 20% to the rentals contracted for as of February 1, 1968 for purposes of arriving at the economic rental value of the property. In his deductions Todd allowed for a vacancy factor of 4%.
Todd, like Olasin, cited the low vacancy experience of Parkview as evidence of inadequate rentals. He pointed out that as an F.H.A. financed project this complex had been subject to government restrictions on rent increases until 1961, when there was a refinancing. He implied that the management never caught up with market rentals thereafter. (Rental income increased from 1959 to 1967 by 16%).
Todd "investigated" the lease-rental status of the property as of February 1968 and testified, over objection, that three *27 leases were scheduled to expire in 1968, 144 in 1969, 472 in 1970 and 373 in 1971. The renewals effected in 1968 averaged 10% rent increases. It was the witness's opinion that as of the assessing date, if the leases did not exist, all of the apartments could have been rented at higher than the then prevailing rental scales "in my opinion, because they are lower than what the standard rental is for a unit of this type". But he alluded to no basis for the "standard rental" mentioned other than the Cherry Hill comparative study.
Todd conceded that in the course of a 1961 appraisal of the Parkview property he had described the management as excellent and attributed the low vacancy experience to that factor. He testified here that that was still his opinion. He was of the view that the Cherry Hill rentals indicated in the study were "proper" for that property in spite of the unusual 10% vacancy rate there. He conceded the low turnover rate characteristic of the Collingswood area would contribute to the low Parkview vacancy experience.
The whole of the consideration and findings by the Division as to the matter of economic rentals will be found in the following portion of its opinion:
"In the economic approach, the expert's values are:
Bate 7,318,219[2]
Olasin 10,280,000
Todd 10,200,000
There was only one serious difference. The respondent's experts both testified that the actual rents were 20% too low and the economic rent should be placed at 20% higher than the actual rent. They seek to support this by a comparison with the Cherry Hill Apartments which are luxury apartments twelve stories high. However, these have central air conditioning and also have a swimming pool. Mr. Olasin testified that, over all the Cherry Hill Apartments are about 5% better. His analysis is set forth in Exhibit R-5. Cherry Hill has a vacancy factor of 10%; the case being tried has about 1 1/2% vacancy record. This would tend to show that the rents in *28 this case are too low. Further, as the leases in the Collingswood complex came due in 1968, the rents were raised 10% and the vacancy factor did not change. This shows the rental income could stand a substantial raise. I find that the actual rent of $1,622,754 should be raised 15% to $1,866,166 as the true economic rent."
Certain observations as to the foregoing come to mind at once. There is no express finding as to the validity of the conclusions of respondent's experts as to the 20% appreciation factor for Parkview based on the Cherry Hill comparison, and one therefore cannot know whether the Division's 15% appreciation factor rests on those conclusions to any extent or only on its other reasons expressed. The comparison was challenged by petitioner at the hearing in a number of respects, but the opinion lacks adequate finding thereon, and we are correspondingly handicapped in our review function. See Rek Investment Co. v. Newark, 80 N.J. Super. 552, 562 (App. Div. 1963) and cases cited therein.
The bare averment in the opinion that the Cherry Hill vacancy of 10% and the Parkview vacancy of 1 1/2% tend to show the Parkview rents are too low is difficult to understand on the record before us. The judge himself used a 3% vacancy factor for purposes of capitalization, as did petitioner's expert, while respondent's experts used 4% and 5% respectively. All of these estimates are much closer to the Parkview experience of 1 1/2% than to that of Cherry Hill at 10%. Yet respondent's experts assumed the Cherry Hill rentals were not too high notwithstanding its vacancy rate, and the judge himself was apparently in accord if he may be understood as implicitly accepting the Cherry Hill comparative study. Moreover, there are the factors of favorable management at Parkview and low-turnover Collingswood pattern to support the consistency of the Parkview vacancy rate with the fairness of its contract rentals. The opinion does not comment on these factors.
Equally questionable is the Division's factual premise of an absence of change in vacancy rate after imposition of *29 10% rental increases in leases coming due in 1968 as a basis for its conclusion that "the rental income could stand a substantial raise." The proofs were that only a handful of leases expired in 1968 and that these were raised for old tenants by 10%, new tenants being charged somewhat more. The fact that there was no immediate change in vacancy rate would therefore seem of little or no relevance to the economic adequacy of the rentals as of October 1, 1967.[3] In any case it would seem that the average tenant could be expected to accept a 10% increase for a renewal in a rising rental market after a lease period of two or three years. The question here is the fair rental value, looking forward as of October 1, 1967, not 1968, 1969 or thereafter. Hindsight as to later leasing arrangements is irrelevant. New Brunswick v. State of N.J. Div. of Tax Appeals, supra (39 N.J. at 545).
There remains for consideration the probative weight of the opinions of respondent's experts so far as based on the Cherry Hill comparative analysis. Our discussion thereof will be pertinent (a) if the Division can be thought to have relied on it; (b) so far as respondents seeks to support the judgment on the basis of it independent of the Division's reasoning; and (c) to guide the parties and the Division at the rehearing ordered in this opinion.
It is of course settled that gross rental income for purposes of applying the capitalized income approach to valuation of property is to be taken at "fair rental value," professionally termed "economic" rent or income, if that differs from current actual rental. New Brunswick v. State of N.J. *30 Div. of Tax Appeals, supra (39 N.J. at 544); American Institute of Real Estate Appraisers, The Appraisal of Real Estate (1967), p. 227 et seq.; International Association of Assessing Officers, Assessing and the Appraisal Process (4 ed. 1972), pp. 82, 172; Kahn, Case, Schimmel, Real Estate Appraisal and Investment (1963), p. 103 et seq.; Ring, The Valuation of Real Estate (1970), p. 208. However, actual income is a significant probative factor in the inquiry as to economic income. See McCrory Stores Corp. v. Asbury Park, 89 N.J. Super. 234, 243 (App. Div. 1965); Somers v. City of Meriden, 119 Conn. 5, 174 A. 184, 186 (Sup. Ct. 1934); People ex rel. Gale v. Tax Commission of City of New York, 17 A.D.2d 225, 233 N.Y.2d 501, 506-507 (App. Div. 1962), motion for leave to appeal denied, 12 N.Y.2d 646, 238 N.Y.S.2d 1026 (Ct. App. 1963); Assessing and the Appraisal Process, supra, at 82; Real Property Appraisal Manual for New Jersey Assessors (2 ed. 1963), p. 148.
Checking actual income to determine whether it reflects economic income is a process of sound appraisal judgment applied to rentals currently being charged for comparable facilities in the competitive area. The essential, however, is a plurality of comparables, see The Appraisal of Real Estate, supra, at 235; Assessing and the Appraisal Process, supra, at 82, 172; Kahn, Case, Schimmel, supra, at 105. One authority suggests that if at least four comparable properties are properly analyzed, a basis for a reliable rent schedule for the subject apartment house may be made. Ring, supra, at 215. The reason for requiring a reasonable number of comparables is obvious. Postulating that the economic rent for property A is X dollars per unit (rather than X minus Y, which A is now charging) on the basis of the mere fact that assumedly similar property B is charging X dollars simply begs the question as to which of the two, if either, is charging economic rent. What is missing is a sufficient sampling of comparable area rents to provide a reliable criterion. The argued unavailability of a sufficient number of comparables cannot overcome the basic defect of relying on only one. (But *31 there was testimony that many high rise apartments have been erected in the Camden area in the last fifteen years.) And that brings us to one of the fundamental deficiencies of respondent's case as presented below so far as it rests on the hypothesis that the Cherry Hill comparative analysis proves Parkview's rents were too low.
We assume for the present discussion (but see infra) that respondent has satisfactorily established that the Cherry Hill apartments should command an economic rent only 5% greater than Parkview and that in fact its rental scale exceeds Parkview's by considerably more than that. There is no particular evidence in this record, nor any finding by the Division, that the Cherry Hill rent schedule is necessarily the economic rent for that property beyond the simple, unexpounded ipse dixit of Messrs. Olasin and Todd (having in mind for whatever it may be worth, among other things, the high 10% vacancy experience of that property.)
We next consider the mechanics of the Olasin comparison study. A number of disturbing questions are either argued by petitioner or arise in our minds independently. These tend to impair the validity of the assumptions we made arguendo in the preceding paragraph.
1. The comparison was done on the basis of unit rent per square foot of apartment space rather than, as customary for apartment houses, per room or per apartment unit. See The Appraisal of Real Estate, supra, at 233, 234; Kahn, Case, Schimmel, supra, at 103, 104, 106; Friedman, Encyclopedia of Real Estate Appraising (rev. ed. 1968), p. 205; Ring, supra, at 215. Compare the situation as to commercial and industrial property, where square foot unit rentals are commonly used in appraisals. See the technical works, ubi supra. The Division judge commented at trial that he had never seen square foot units used in apartment house appraisals. Respondent's experts, though conceding they had never employed the method before, defended it as reducing error because *32 entailing use of a "least common denominator."[4] One wonders what the analysis would have shown if based on comparison of units of rental value by room or apartment-room-number, as customary.
2. A substantial discrepancy arose as to the Parkview measurements submitted by Olasin. The seven sample apartments used were measured by him personally. But petitioner apparently showed that the actual floor plans of the apartments in question as filed in the municipal building department were less in area by about 25% than Olasin's measurements. There was no finding of fact by the Division as to this critical conflict. Olasin curiously used filed building plans for his Cherry Hill dimensions as contrasted with personal measurements at Parkview.
3. As already noted, Olasin disregarded the fact that while rentals were higher on higher floors at both properties the Cherry Hill apartments were 2 to 3 stories higher than Parkview, and he made no adjustment therefor.
4. The fairness of the sampling of apartments is suspect. Out of over 1,000 apartments at Parkview Olasin constructed his square foot unit value on the basis of only seven apartments, six of them in one of the four buildings. At Cherry Hill, by contrast, he selected 36 units, or eight per cent of the total. In each case the witness said the sampling was representative "of the mix" and comparable, but he did not give supporting detail for the conclusions. The exhibits in evidence appear to indicate to the contrary.
Whereas Parkview's apartment types run in the following approximate percentages of the whole:
3 1/2 rms. 58% 4 1/2 rms. 13% 5 1/2 rms. 29%*33 Olasin's sample "cross-section" was of two 3 1/2 room apartments (28%), two of 4 1/2 rooms (28%) and three of 5 1/2 rooms (44%) an obvious overrepresentation of the 5's and 4's and underrepresentation of the 3's. The prejudice of the disproportion arises from the fact that larger apartment units rent for less per room (and therefore also per square foot) than smaller ones at both Parkview and Cherry Hill. It is generally the case that rents of larger apartments reduce to lesser area unit rates than smaller ones; see Kahn, Case, Schimmel, supra, at 106. Thus it would appear that Olasin's sampling method produces an artificially deflated square foot rental unit for Parkview.
Moreover, assuming Olasin's sampling schedule of 36 units at Cherry Hill is representative of the mix of the whole (not determinable from the record) one finds an apparent substantial difference in the apartment size categories at Cherry Hill compared with Olasin's Parkview mix. Olasin's Cherry Hill sample discloses approximately 28% of 3-room units, 50% of 4-room and 22% of 5-room.[5] This difference from Olasin's mix at Parkview (not to mention the actual Parkview mix) may be productive of additional distortion in the comparison of square foot unit rentals by virtue of the same considerations as mentioned in the preceding paragraph. If there are answers to the difficulties we here express they can be provided at the rehearing.
5. At both complexes leases varied in expiration date from one to three years. Earlier leases were inferentially at lower scales than later ones, at least to some extent. Olasin made no attempt to coordinate comparative sample selections in relation to age of lease.
6. Of major concern is the seemingly simplistic approach Olasin took to the inherently complex but crucial problem of comparing Cherry Hill and Parkview from a market rental value standpoint. He used only four comparison category *34 indices, lumping separate significant features within them. As noted above, he assigned comparison ratios to three of the indices in round multiples of 5%, these being 5%, 10% and 10%, and found the fourth index to be equal.[6]
We thus conclude from everything we have said that the determination of the Division that 15% should be added to Parkview's contract rentals to arrive at its fair rental value as of October 1, 1967 is not supported by substantial credible evidence on the whole record, allowing for agency expertise and evaluation of the credibility of witnesses.[7] Moreover, the determination of the Division is, as noted, conspicuously deficient in essential findings of fact in relation to the matters we have discussed. In the absence of convincing evidence to the contrary the current ongoing income scale of a large, well-managed apartment project like this, functioning as customary with leases of relatively short length, should be deemed prima facie to represent its fair rental value for purposes of the capitalized income method of property valuation. A court or taxing agency should be most hesitant to find that the tenants of a residential property being operated commercially are being charged inadequate rent. That approach, we believe, conduces *35 to the objective of relative stability of assessments which we have heretofore held to be basic to sound tax assessment policy. New Brunswick v. State of N.J. Div. of Tax Appeals, supra (39 N.J. at 541); Switz v. Middletown Twp., 23 N.J. 580, 596 (1957); Hackensack Water Co. v. Division of Tax Appeals, 2 N.J. 157, 163 (1949). Readily to be distinguished is the case of a taxpayer owning commercial property tied to a long term lease made long before the current assessing date, where the present rent may well be out of line with current fair rental value. See New Brunswick, supra (39 N.J. at 544).
The petitioner calls our attention to certain other alleged defects in the Division's capitalization of income. One of them, of seeming merit, is the use of a figure for actual apartment rentals as of October 1, 1967 (or February 1, 1968), prior to application of the 15% appreciation factor, higher than any figure therefor in evidence. We think these matters may become academic in the light of what we have said herein and a redetermination of the appeal. In any event, the points raised may be argued before the Division.
Reversed and remanded to the Division of Tax Appeals for a new hearing and determination, to be accompanied by adequate findings of fact and conclusions on all relevant issues.
For reversal and remandment Chief Justice WEINTRAUB, Justices JACOBS, HALL and MOUNTAIN, and Judges CONFORD and SULLIVAN 6.
For affirmance None.
NOTES
[1] Both of respondent's experts testified that Parkview's management was good. Yet Olasin did not appear to consider whether that factor might have contributed to the low vacancy result. Compare the admissions on this point by respondent's other expert, Todd, mentioned hereinafter.
[2] This figure is erroneous. See supra. It represents only the improvement valuation on the income approach.
[3] Petitioner filed affidavits with the Appellate Division in support of a request for a rehearing on grounds of newly discovered evidence. These purport to show that when rents of several hundred apartments were increased substantially at one time in 1970 vacancies rose to 13%. While we do not rely on this data for our determination of this appeal since it is not properly in the record and shows facts long after the assessing date, it tends if true to disparage the Division's reasoning in the portion of its opinion just alluded to.
[4] It has been suggested that square foot estimates "for several buildings" can be used as a check against the primary indications from per room rentals if questions arise as to half rooms. Kahn, Case, Schimmel, supra, at 104.
[5] We assume "I Bedrm., 2 Bedrm.", etc. in the schedule mean, respectively, 3-room and 4-room apartments.
[6] Compare Kahn, Case, Schimmel, supra, at 104, listing "some of the most important factors" to consider, e.g., size of rooms, age of building, number of rooms per apartment, appointments (interior finish, kitchen, bath) layout and exposure, location of apartment (upper or lower floor), air conditioning, prestige address, surroundings, services received.
[7] We see no reason why this standard of judicial review, which we have applied in recent years to determinations of administrative agencies generally, see, e.g., Atkinson v. Parsekian, 37 N.J. 143, 149 (1962); Close v. Kordulak Bros., 44 N.J. 589, 599 (1965); Passaic v. Gera Mills, 55 N.J. Super. 73, 92 (App. Div. 1959), should not also apply to review of the Division of Tax Appeals. Compare the variant formula previously used in tax cases. Aetna Life Insurance Co. v. City of Newark, 10 N.J. 99, 104 (1952); In re West New York, 25 N.J. 377, 385 (1957); Weston Electrical Instrument Corp. v. Newark, 11 N.J. Super. 493, 496 (App. Div. 1951).