These consolidated local property tax cases concern appeals from judgments of the Bergen County Board of Taxation affirming all assessments for the years 1983 through 1987 on finished and partially finished townhouse condominium units comprising part of the condominium development known as The Courts of Glenpointe in Teaneck, New Jersey. Plaintiff, Glenpointe Associates, also seeks review of county board judgments affirming the 1985, 1986 and 1987 assessments on the clubhouse, tennis courts, swimming pool and gatehouse located at the condominium development and constituting a common element as that term is defined in N.J.S.A. 46:8B-3(d). The common element, identified as Block 3712, Lot 5-C 0000, was assessed for all relevant years at:
Land $ -0-
Improvements 269,000
Total $269,000
The condominium, created by master deed filed with the Bergen County Clerk’s office on July 23, 1982, covers 14.471 acres, is part of Block 3712, Lot 5, and comprehends 171 units constructed, or to be constructed, in two phases. Phase I consists of 88 units, all of which were finished and ready for occupancy by *292December 1982. The 83 units comprising Phase II were in various stages of completion on the assessing dates but none was completed and ready for use by the last assessing date, i.e., October 1, 1986.
Not all 171 units are under appeal. The units which are the subject of this proceeding are as follows:
Phase I Phase II
1983 76 83
1984 69 83
1985 87 83
1986 88 83
1987 88 73
The units are all individually assessed in accordance with N.J.S.A. 46:8B-19. Glenpointe Associates, the condominium sponsor, is plaintiff in most of the appeals, but many appellants are owners who purchased their units from the sponsor.
The finished units have been valued by both experts by means of the market data approach; the unfinished units have been valued by the cost approach, supplemented by the market data approach for the land value.
I.
The recreational facilities.
The court will first address the assessment of the common element consisting of the clubhouse, tennis courts, swimming pool and gatehouse (hereinafter sometimes referred to as the recreational facilities).
The issue with regard to the recreational facilities is whether they may be separately assessed or whether the assessments of the condominium units must include a proportionate share of such facilities, thereby eliminating the separate assessment on those facilities. The issue arises because,' on the assessing dates for the years 1985, 1986 and 1987, only 88 of 292 contemplated townhouse condominium units were completed and ready for occupancy, whereas the recreational facilities *293were completed and ready for use on all the relevant assessing dates.
The parties agree that the recreational facilities constitute a common element as that term is defined in N.J.S.A. 46:8B-3(d).
The courts of Glenpointe was established as a condominium by the filing of a master deed with the Bergen County Clerk’s office on July 23, 1982. The condominium covers 14.471 acres and is part of Block 3712, Lot 5. The master deed and the offering plan contemplated the immediate development of regime “A” consisting of 171 townhouse units. The master deed and offering plan also revealed the sponsor’s intention to establish regime “B” at some future time, which would comprise an additional 121 condominium units. All the units of both regimes would be served by the common elements represented by the recreational facilities.
The master deed provided that the owner of each unit in regime “A” also owned a percentage interest in the common elements. That interest was indivisible from the unit to which it pertained. Construction of the first 88 units was complete by the end of 1982. No additional units were completed and ready for occupancy before October 1, 1986.
The disposition of the issue before the court is governed by the Condominium Act, N.J.S.A. 46:8B-1 et seq.
A condominium is created and established by recording a master deed with the recording officer of the county in which the land is situated. N.J.S.A. 46:8B-8. Each condominium unit constitutes a separate parcel of real property with which the owner may deal in the same manner as the law allows with respect to any other parcel of real property. N.J.S.A. 46:8B-4. The proportionate undivided interest in the common elements assigned to each unit is inseparable from such unit, and any conveyance, lease, mortgage or other disposition of a unit extends to the proportionate share of the common elements assigned to such units. N.J.S.A. 46:8B-6.
*294All property taxes are separately assessed against each unit as a single parcel and not on the condominium property as a whole. N.J.S.A. 46:8B-19. The term “unit” is defined to include the share of the common elements appurtenant to it. N.J.S.A. 46:8B-3(o). The separate real property tax assessment for each condominium unit facilitates the goal of the Condominium Act to constitute each unit as a separate parcel of real property, to be dealt with by the owner in the same manner as any other parcel of real property. Troy Village Realty Co. v. Springfield Tp., 191 N.J.Super. 559, 468 A.2d 445 (App.Div.1983).
Defendant contends that only 30% of the recreational facilities are relieved from separate assessment. Its theory is that, as only 88 units of a contemplated 292, i.e., 30% of the total number of units contemplated by the condominium development, were completed and ready for use on the relevant assessing dates, the recreational facilities in question, which were constructed and ready for use prior to the first assessing date, served only those 88 units.
The flaw in this argument lies in the unstated assumption that the value of the common elements (i.e., the recreational facilities) can only be transferred to, and included in, the value of completed condominium units. The Condominium Act imposes no such limitation. By statute, a condominium is created and established upon the filing of a master deed. The statute also clearly provides that each condominium unit is separately assessed and that, appurtenant to and inseparable from each unit, is a proportionate share of the common elements. Indeed, the statutory definition of a condominium unit includes the share of the common elements assigned to it. The separate assessment of each unit contemplated by N.J.S.A. 46:8B-19 may involve land only, or it may reflect land plus partial improvements, such as footings and foundations. Whatever the condition of the unit assessed, whether finished or unfinished, the unit’s assigned share of the value of the common elements is included in the assessment.
*295The separate assessment of the recreational facilities for the years 1985, 1986 and 1987 will be eliminated.
Plaintiff argues that it is entitled to the same relief with respect to the common elements in question for tax year 1984. No complaint was filed with this court for that year and the record does not indicate whether an appeal was filed with the Bergen County Board of Taxation for 1984. Plaintiff bases its claim for relief upon the statute dealing with duplicative assessments, N.J.S.A. 54:4-54, which provides for corrective action by a municipal governing body or county board where real property is erroneously entered and assessed twice on the tax records. Plaintiff also asks this court to invoke its equity jurisdiction pursuant to N.J.S.A. 2A:3A-4(a). That statute authorizes this court to grant legal and equitable relief in order that all matters in controversy may be completely determined.
The jurisdiction of this court in local property tax matters includes review of county board actions pursuant to N.J.S.A. 54:51A-l(a) and direct review of large assessments pursuant to N.J.S.A. 54:3-21. The time for filing a complaint with this court under N.J.S.A. 54:51A-l(a) with respect to 1984 has long since expired. Accordingly, as plaintiff failed to file a timely complaint with this court with respect to 1984 the court is without jurisdiction to award plaintiff relief. The application of N.J.S.A. 2A:3A-4(a) concerning this court’s equity jurisdiction presupposes jurisdiction of the subject matter in the first instance.
As indicated above, the condominium development known as The Courts of Glenpointe comprehends 171 units constructed, or to be constructed, in two phases.
II.
The finished units.
Eighty-eight townhouse condominiums were built by October 1, 1982. Their locations are as follows:
*296Building No. Address Number of Units
1 Lawrence Court 13
6 Millay Court 13
7 Lowell Court 18
8 Sandberg Court 18
9 Eliot Court 13
12 Whitman Court 13 88
Sales of the units occurred as follows:
Number of Sales Year
10 1982
6 1983
14 1984
35 1985
14 1986
79
The townhouse condominiums consisted of nine models of varying sizes, numbers of bedrooms and amenities. Some were one-story models, others were two-story models. Each building contained at least one of each of the nine models. The 79 sales which occurred in the years 1982 through 1986 involved all nine models, but the sales disclosed no pattern of price or range of prices according to model type. The sales prices of model A-3 units, for example (two-story, two-bedroom, 1,052 square feet), ranged from $114,000 on November 22, 1983 to $186,000 on April 24,1986. A model A-3 unit sold for $140,000 on January 10, 1984, less than two months after another A-3 unit in a different building sold for $114,000. A model A-3 unit sold for $180,000 on July 18, 1986, three months after another unit of the same model but in a different building sold for $186,000.
An even more dramatic demonstration of the fact that the market rejected a price stratification according to unit model is found in the sales of an E-2 model unit and an F-l model unit. The F-l model has a larger living area (1,922 square feet compared to 1,908 for the E-2) and has three bedrooms compared to two for the E-2. The E-2 is a two-story unit with a balcony; the F-l is a one-story unit with a patio. Yet an F-l *297unit sold on April 27,1984 for $205,000, while an E-2 unit in the same building as the F-l sold less than two weeks later for $230,000.
Plaintiffs expert valued the units according to model and square feet of living area. His value estimate was based upon sales of the models, adjusted for reduction of items of personalty included in the sale, specifically, refrigerator, washer and dryer. He also adjusted the sale price for tax years 1983 and 1984 to account for what he believed to be an extended marketing period, which was two years for units sold in 1982 and six months for units sold in 1983. This resulted in a 25% discount for 1982 (for tax year 1983) and 12% for 1983 (for tax year 1984).
The expert reasoned that high interest rates in 1982 and 1983 discouraged buyers and adversely affected the market’s absorption of the 88 finished units. He points to the relative dearth of sales in 1982 and 1983 and the sharp increase in sales in 1984, 1985 and 1986 when interest rates dropped.
Defendant’s expert also used the comparable sales approach, relying upon sales of the units themselves to value not only the sold units but the units still owned by Glenpointe Associates. He valued the sold units essentially at their selling prices in estimating values for the assessment year and adjusted those values by 10% to arrive at value estimates for subsequent and, where appropriate, prior years.
The expert declined to value the units according to models. He concluded that location in a given building, the location of the building itself, and the view afforded a particular unit made such an approach unreliable. Rather, he regarded the selling price as a more dependable indicator of market action.
The testimony and appraisal of plaintiff’s expert are not probative of the true value of the subject properties. The probative value of an expert’s opinion depends entirely upon the facts and reasoning offered in support of it. Dworman v. Tinton Falls, 1 N.J.Tax 445 (Tax Ct.1980), aff’d o.b. per cu*298riam 180 N.J.Super. 366, 3 N.J.Tax 1, 434 A.2d 1134 (App.Div.1981).
The expert’s adjustment for the washer, dryer and refrigerator is not supported by the record. While the expert theorized that these items could be removed without damage to the unit and that they are not ordinarily sold with the real estate, he admitted that he had not examined any of the contracts of sale to verify his suppositions.
The expert’s marketing discount theory must be rejected as a matter of law, as it violates the uniformity clause of the New Jersey Constitution. The decision of this court in Tall Timbers, Inc. v. Vernon Tp., 5 N.J.Tax 299 (Tax Ct.1983) is directly in point. That case involved a condominium campsite which had been subdivided into 534 separately assessed parcels. In valuing the campsites for assessment purposes the taxpayer’s appraiser discounted the average selling price by 50% to account for overhead and marketing costs. The taxing district’s appraiser, on the other hand, discounted the value of the unsold campsites which remained in the hands of the developer while valuing the sold campsites at their full selling prices. Judge Lasser rejected the discount theories of both experts and held that each campsite must be assessed at its market value as reflected by comparable sales without discounting. While recognizing the existence of appraisal authority to support a developer’s discount for marketing a subdivision, the court held that, for local property tax purposes, each condominium campsite must be assessed separately “according to the same standard of value.” N.J. Const. (1947), Art. VIII, § 1, par. 1(a).
After analyzing the history of that section of the Condominium Act governing real property tax assessments (N.J.S.A. 46:8B-19), Judge Lasser concluded that application of the discount theory would violate the uniformity clause. His reasoning is worth quoting:
In the subject case, uniformity of assessment requires the assessment of each campsite to be based on its market value as reflected by comparable sales, without discounting. Marketing costs and time considerations are reflected in the sales price of each campsite. There is no justification for discounting the *299sales price of a campsite that has been sold. To value a sold campsite at a discounted value would be to use a “standard of value” different from that used for all other properties in the taxing district, which are valued based on their market value. To permit discounting of the sales prices of the sold campsites would result in a lack of uniformity between these campsite assessments and other assessments in the taxing district. Uniformity of assessment requires separate valuation of each campsite, whether one person owns one or many. [5 N.J.Tax at 307]
Moreover, the expert’s reflection of temporary high interest rates in his value estimates is contrary to the principle of assessment stability, which holds that valuation for tax assessment purposes must have some measure of permanency rendering it secure against general temporary conditions such as inflation or deflation. Hackensack Water Co. v. Div. of Tax Appeals, 2 N.J. 157, 163, 65 A.2d 828 (1949); Rek Investment Co. v. Newark, 80 N.J. Super. 552, 559, 194 A.2d 368 (App.Div.1963). As we said in Inwood at Great Notch v. Little Falls Tp., 6 N.J.Tax 316 (Tax Ct.1984) with specific reference to high rates:
The strong public interest in assessment stability and its corollary, stable, predictable municipal revenues, require that tax assessments not be hostage to sharp fluctuations traceable to volatile swings in the financial markets____ [6 N.J.Tax at 332]
Thus, the principle of assessment stability requires that temporary high interest rates, which tend to extend the marketing period for all properties in the municipality, be disregarded.
This court has also ruled that the time required to market a property must be disregarded for local property tax purposes. ITT Cont’l Baking Co. v. East Brunswick Tp., 1 N.J.Tax 244 (Tax Ct.1980). In that case Judge Andrew observed:
The value to be determined is as of October 1 of the pretax year, a specified date. N.J.S.A. 54:4-23. The question is, what would the property sell for at a bona fide sale on October 1? Ibid. The period of time it would take to sell realty does not enter into a determination of value as of the specified date____ [Id. at 250]
The conclusions of defendant’s expert are supported by sound reasoning and by the credible evidence. Unlike plaintiff’s expert he declined to value the condominium units according to models, as the sales conclusively demonstrated the ab*300sence óf a significant relationship between models and selling prices. Defendant’s expert wisely let the market do its work and his value estimates are accurate reflections of market action. His 10% annual adjustment for time is, if anything, conservative. The evidence in this case indicates a large number of resales of the subject units; and those resales reflect value increases, measured by sale prices, of anywhere from 9.8% annually to as much as 28% annually.
In view of the foregoing I find the true value of all finished condominium units (Phase I) to be as estimated by defendant’s expert.
III.
The unfinished units.
Eighty-three townhouse condominiums units (Phase II) were under construction on all relevant assessing dates. Their locations are as follows:
Building Number of
No. Address Units
2 London Court 13
3 Alcott Court 13
4 Carlyle Court 18
5 Wadsworth Court 13
10 Peabody Court 13
11 Sinclair Court 13
83
Plaintiff’s expert valued each unfinished unit based on actual construction costs furnished by plaintiff, to which he added an allocable land value based upon plaintiff’s land acquisition cost pursuant to a 1977 contract with Teaneck Redevelopment Agency. In estimating the value of the improvements, i.e., the footings and foundations, he did not add contractor’s profit. While he claimed to use actual construction costs he testified that he trended the cost for each year after the first tax year (1983) according to the Marshall Valuation cost index. Finally, *301he made an adjustment for depreciation of 1% per annum.1 He thus arrived at the following value estimates for each of the Phase II units (including land):
Buildings 2, 3, 5, 10, 11
Year Costs Time Factor2 Depreciation Value
10/1/82 $10,308 1.00 0 $10,300
10/1/83 10.308 1.086 (100) 11,100
10/1/84 10.308 1.178 (200) 11,900
10/1/85 10.308 1.257 (400) 12,600
10/1/86 10.308 1.331 (500) 13,200
Building 4
Year Costs Time Factor2 Depreciation Value
10/1/82 $ 8,889 1.00 0 $ 8,900
10/1/83 8.889 1.086 (100) 9,600
10/1/84 8.889 1.178 (200) 10.300
10/1/85 8.889 1.257 (300) 10,900
10/1/86 8.889 1.331 (500) 11.300
Defendant’s expert valued each of the Phase II units on the basis of the actual cost of construction, plus overhead and contractor’s profit, as of each assessing date. He assigned a land value of $16,500 a unit for tax year 1983 and adjusted this estimate by approximately 10% for each subsequent year. His land value estimate was predicated on nine comparable sales of land purchased for condominium development. In addition to the total sale price and date of sale he indicated the total acreage, permissible density in terms of units per acre and the price per unit for each sale. He testified that developers buy land for condominium use on the basis of permissible density, i.e., they are willing to pay for the land according to the maximum number of condominium units allowed per acre. The *302prices paid by the purchasers of the nine parcels of land, expressed in terms of condominium units, ranged from $4,700 a unit per acre to $29,800 a unit per acre. For seven of the nine sales the prices per condominium unit ranged from $10,000 to $23,500. The dates of the sales ranged from June 1980 to April 1984, all in reasonable proximity to the first three assessing dates. The units per acre to be constructed on the sites of the comparable land sales ranged from 3.1 to 27. The most comparable of these sales to the subject are sales # 1 and # 3, which involve, respectively, 71 and 120 units, 15.7 acres and 8.12 acres and 4.4 units per acre and 14.8 units per acre. The total number of units planned for both Phase I and Phase II of the subject condominium development is 171; the total land area of the subject is 14.471. Thus, 11.8 units per acre were contemplated by the developer at the time of acquisition of the subject land. The relevant Teaneck Ordinance permits 15 units an acre.
The expert valued each of the Phase II units as follows:
1983 1984 1985 1986 1987
Land $16,500 $18,000 $20,000 $22,000 $24,000
Impvts. 7,500 7,900 8,300 8,700 9,200
Total $24,000 $33,200 $25,900 $28,300 $30,700
Defendant’s expert also estimated the true value of the clubhouse, gatehouse, swimming pool and tennis courts (the recreational facilities) as of the assessing dates for tax years 1984 through 1987. Using construction cost data supplied by plaintiff he estimated the true value of the recreational facilities to be $337,900 on October 1, 1983. He then adjusted this amount by 10% annually to arrive at the following value estimates for the ensuing years:
10/1/84 $371,700
10/1/85 408,900
10/1/86 449,800
*303Plaintiff’s expert declined to value the recreational facilities on the ground that, as common elements of a condominium development, they could not be separately assessed.
The valuation estimates of plaintiff’s expert are critically flawed. First of all, his allocation of land value, based as it is upon plaintiff’s land acquisition cost under a 1977 contract, is not probative of the subject property’s true value on October 1, 1982, the earliest of the assessing dates in issue. The contract was executed at a point which is too remote in time to provide a reliable indication of value. The expert opined that interest rates, construction costs and the costs of on-site and off-site improvements all increased in the interval between the date of execution of the contract and the assessing date in question. However, he admitted that increased costs and interest rates affected all vacant land in Bergen County; yet he made no study of vacant land sales to prove the absence of appreciation in land values during the time interval in question. As for the increased cost of on-site and off-site improvements he opined that, if the subject property had been filled, graded and ready for construction when the purchase contract was signed, he would have made no time adjustment. Thus, the rising cost of on-site and off-site improvements could not have been a material factor in the expert’s decision to make no time adjustment.
Moreover, his estimates of improvement values are not convincing. He claimed to use actual costs in accordance with data furnished by plaintiff but, from his appraisal and testimony, it appears that he used actual costs for 1982 only, and trended those costs for the remaining years according to the Marshall Valuation cost index. Actual cost data were available for later years and they should have been used.
The value estimates of defendant’s expert, on the other hand, are well documented and otherwise amply supported by the credible evidence. His comparable land sales support his land value estimate; in particular they demonstrate the accuracy of his observation that condominium developers frequently buy land on the basis of anticipated density of condominium *304units. His time adjustment of 10% annually is also reasonable and, the evidence shows, is somewhat on the conservative side. His allocation of a uniform land value to each unit is also reasonable. The evidence submitted with respect to values of the finished units indicates that differences in values of the units are ascribable more to the view and amenities of the improvements than to the underlying land.
Furthermore, while the master deed sets forth the percentage interest in the common elements (which includes the land) for each of the 171 units broken down by buildings and the percentage interests appear to relate to the nine models, there is no explanation in the master deed or the offering plan of the quantum of differences in the percentages among the models. It is not for this court to speculate on the basis for the differences.
Likewise, the expert’s value estimates of the improvements are soundly based upon construction costs actually incurred as at each assessing date. Original cost of construction is properly to be considered on the question of value for property tax purposes. Bostian v. Franklin State Bank, 167 N.J.Super. 564, 401 A.2d 549 (App.Div.1979), on remand 1 N.J.Tax 270 (Tax Ct.), aff’d per o.b. per curiam 179 N.J.Super. 174, 2 N.J.Tax 391, 430 A.2d 1140 (App.Div.1980). In this case there are no other proofs of value of the pilings and foundations in Phase II; construction cost is thus controlling. The expert was also correct in adding contractor’s profit as plaintiff acted as its own general contractor. Sound appraisal principles require the inclusion of contractor’s profit in direct costs. American Institute of Real Estate Appraisers, The Appraisal of Real Estate (9 ed. 1987) 352.
The land is a common element, N.J.S.A. 46:8B-3(d)(i); so are the recreational facilities. The proportionate share of the common elements assigned to each condominium unit is inseparable from such unit and any conveyance, lease, mortgage or other disposition of a unit extends to the unit’s proportionate share of the common elements. N.J.S.A. 46:8B-6. All property taxes *305must be separately assessed against each unit as a single parcel. N.J.S.A. 46:8B-19. The definition of unit includes the proportionate share of the common elements assigned to it. N.J.S.A. 46:8B-3(o).
Thus, the common elements may not be separately assessed; a proportionate share of their value, however, is included in the tax valuation of each unit. This value includes the land value assigned to each unit and a proportionate share of the value of the recreational facilities.3 The court finds those values to be as estimated by defendant’s expert for all the years in issue.
The land value ascribable to each unit is:
1983 $16,500
1984 18,000
1985 20,000
1986 22,000
1987 24,000
The proportionate share of the value of the recreational facilities will be added to the land value and the cost-derived improvement value in the following manner:
1984 $337,900 ~ 171 units = $1,980*
1985 371,700 -h- 171 units = 2,170
1986 408,900 + 171 units = 2,390
1987 449,800 -h 171 units = 2,630
* All figures in this column are rounded.
In view of the foregoing I find the true value of each unit in Phase II to be as follows on the indicated assessing dates:
*30610/1/82 10/1/83 10/1/84 10/1/85 10/1/86
Land $16,500 $18,000 $20,000 $22,000 $24,000
Improvements 7,500 7,900 8,300 8,700 9,200
Allocable share of rec’l facilities -0-1,980 2,170 2,390 2,630
True value $24,000 $27,880 $30,470 $33,090 $35,830
IV.
Discrimination.
Before addressing the application of chapter 123 the court must deal with plaintiff’s claim that the statutory remedy for discrimination relief is inadequate and the court is required by the New Jersey Constitution to apply the unweighted, unclassified ratio to its finding of true value for each year in order to fix the assessments on the subject property.
Our Supreme Court has declared that chapter 123 is the exclusive remedy for assessment discrimination relief, unless it is shown that the application of the statute leaves the taxpayer with a virtually confiscatory assessment. Murnick v. Asbury Park, 95 N.J. 452, 471 A.2d 1196 (1984) (citing with approval 525 Realty Holding Co. v. Hasbrouck Heights, 3 N.J.Tax 206 (Tax Ct.1981) and other decisions of this court). The Supreme Court also observed that “a taxpayer’s right to relief should be determined in accordance with chapter 123 in all but the most extreme or severe circumstances.” 95 N.J. at 463, 471 A.2d 1196.
In an effort to prove the existence of those extreme and severe circumstances — and the confiscatory nature of the application of chapter 123 — plaintiff offered the testimony of Doctor Arthur Cohen, a statistician. From a statistical sampling of 326 sales Doctor Cohen found the average assessment-to-sales ratio for 1983 to be 62.42%. The chapter 123 ratio for the same year was 71%. Quite apart from the fact that the difference between these two ratios does not demonstrate extreme or severe circumstances or the confiscatory nature of the assess*307ments after applying chapter 123, the comparison of the two ratios is meaningless. Doctor Cohen’s ratio is unweighted and unclassified and covers sales for only one year, whereas the chapter 123 ratio is weighted by property class and year and covers several years. Doctor Cohen’s ratio is derived from sales occurring in the 12 months ended June 30, 1983; the ratio employed in chapter 123 is determined partly on the basis of sales occurring during the 12 months ended June 30, 1982. What this court said in Rudd v. Cranford Tp., 4 N.J.Tax 236 (Tax Ct.1982), one of the decisions cited with approval in Murnick v. Asbury Park, supra, applies with equal force here:
The chapter 123 ratio is promulgated by the Director as of October 1 of the pretax year. N.J.S.A. 54:l-35a. Plaintiff, however, uses the unweighted, unclassified ratio promulgated as of October of the tax year. Plaintiff's comparison is inappropriate. The constitutional sufficiency of the statutory remedy for discrimination can only be examined by comparing two ratios promulgated as of the same date. Given the obvious fact of a significant annual appreciation in value of the generality of properties in a municipality such as Cranford, and the reflection thereof in a declining average ratio, the use of a ratio promulgated one year later than the ratio with which it is compared in aid of ascertaining the latter’s constitutional adequacy is manifestly unfair and inequitable to defendant and its other taxpayers. When the ratios are realigned so that the unweighted ratio and the chapter 123 ratio are viewed at the same temporal reference point, the quantitative difference between the unweighted, unclassified ratio and the chapter 123 ratio is significantly narrowed. It is narrowed, in fact, to the point where such difference loses all constitutional significance.
In addition, the unweighted ratio, derived from a mere one-year sales study, is subject to temporary market fluctuations and ascribes inordinate influence to inadequate sales samples in a single year. The Legislature recognized this when it changed the chapter 123 ratio from a one-year, unweighted and unclassified ratio to the classified and weighted ratio, effective for 1979 and later years. See Committee Statement to Assembly Bill 1492, L.1979, c. 51. The new chapter 123 ratio reflects sales over a period of years, with a different weight assigned to each year on a multiple regression basis. The evident legislative design was to promote stability in assessments, and the statute implementing that objective enjoys a strong presumption of constitutionality____ [4 N.J.Tax at 248-249; citations omitted]
The foregoing analysis applies with equal force to tax years 1985, 1986 and 1987.
*308The application of Doctor Cohen’s sales ratios for tax year 1984 is not feasible, as 1984 was a revaluation year. As Doctor Cohen relied upon sales occurring in the 12 months ended June 30, 1984, the sales occurring in the last six months of that year involved properties that were all revalued at 100%. Thus, there is no basis for comparison of post-1983 sales with sales occurring prior to July 1, 1983, because the latter involved properties assessed at a level substantially below 100%.
For these reasons, then, I conclude that plaintiff has not shown that the application of chapter 123 leaves it with confiscatory assessments for any of the years before the court. Chapter 123 will thus be applied to all years except 1984, which was a revaluation year to which the statute does not apply. As this court has observed on several prior occasions, the application of chapter 123 is automatic; it must be applied where the ratio of assessed to true value exceeds the upper limit or falls below the lower limit of the common level range. Weyerhaeuser Co. v. Closter Boro., 190 N.J.Super. 528, 464 A.2d 1156 (App.Div.1983); Devonshire Development Assoc. v. Hackensack, 184 N.J.Super. 371, 2 N.J.Tax 392, 446 A.2d 201 (Tax Ct.1981).
Counsel will prepare computations, to be exchanged and submitted to the court pursuant to R. 8:9-3, showing the correct assessments for each of the Phase I units under appeal in accordance with the court’s findings of true value herein-above set forth and the application of N.J.S.A. 54:51A-6 for all years under review except 1984, which is a revaluation year to which the statute does not apply. As to that year the court’s finding of true value will be the correct assessment, except where such value exceeds the original assessment, in which case the original assessment will be affirmed. As defendant failed to seek an increase before the county board for 1984 it is foreclosed from such relief in this court, even though, for that year, in any given case, the true value may exceed the assessment. F.M.C. Stores Co. v. Morris Plains Boro., 100 N.J. 418, 495 A.2d 1313 (1985).
*309The average ratios and upper and lower limits of the common level range are as follows:
Year Average ratio Upper limit Lower limit
1983 71% 82% 60%
1985 90.48% 100% 76.91%
1986 81.38% 93.59% 69.17%
1987 67.33% 77.43% 57.23%
The correct assessments for each unit in Phase II for all years under review are as follows (rounded):
1983 1984 1985 1986 1987
Land $11,700 $15,000 $18,000 $17,900 $16,400
Impvts. 5,300 7,000 9,500 9,000 7,660
Total $17,000 $22,000 $27,500 $26,900 $24,000
Judgments will be entered accordingly, following submission of the aforementioned calculations.
Plaintiffs expert erroneously included his allocable land value in the depreciation estimate.
Cost index, Marshall Valuation Service.
Normally, the value of common elements is reflected in sales, but in this case, where we are dealing with the value of unfinished units, the only way to reflect the value of the common elements is to employ the method utilized by defendant's expert.