Decided: October 7, 1985.
J.W. FIELD COMPANY, INC. AND JACK W. FIELD, PLAINTIFFS,
TOWNSHIP OF FRANKLIN, PLANNING BOARD OF TOWNSHIP OF FRANKLIN, FRANKLIN TOWNSHIP SEWERAGE AUTHORITY AND STONY BROOK REGIONAL SEWERAGE AUTHORITY, DEFENDANTS. JZR ASSOCIATES, INC., PLAINTIFF, V. TOWNSHIP OF FRANKLIN ET ALS, DEFENDANT. FLAMA CONSTRUCTION CORPORATION, PLAINTIFF, V. TOWNSHIP OF FRANKLIN ET ALS, DEFENDANT. WOODBROOK DEVELOPMENT CORP., PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANT. WHITESTONE CONSTRUCTION, INC., PLAINTIFF, V. TOWNSHIP OF FRANKLIN ET ALS, DEFENDANT. BRENER ASSOCIATES, PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANT. RAKECO DEVELOPERS, INC, PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANT. JOHN H. VAN CLEEF, SR., JOHN E. VAN CLEEF, JR. AND BONNIE VAN CLEEF, PLAINTIFFS, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANTS. LEO MINDEL, PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANT. R.A.S. LAND DEVELOPMENT COMPANY, INC., PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANT. JOPS COMPANY, PLAINTIFF, V. TOWNSHIP OF FRANKLIN, ET ALS, DEFENDANTS
[206 NJSuper Page 167] This Mount Laurel case provides the court with the opportunity to continue the process of development of a method of fair share allocation as required by Southern Burlington Cty. N.A.A.C.P. v. Mt. Laurel Tp., 92 N.J. 158 (1983). (hereinafter Mount Laurel II) (all page citations shall refer to Mount Laurel II unless otherwise noted) The purpose of the inquiry is to determine the fair share obligation of the Township of Franklin and to reexamine certain aspects of the fair share methodology adopted by this court in AMG Realty Company, et
[206 NJSuper Page 168]
al. v. Warren Tp., 207 N.J. Super. 388 (Law Div.1984) (hereinafter AMG) All municipalities are required to provide a realistic opportunity for the construction of their fair share of the region's present lower income housing need generated by dilapidated or overcrowded units, including their own. Municipalities like Franklin, which contain "growth areas" as shown on the State Development Guide Plan, (hereinafter SDGP) also have an obligation to provide for their fair share of the region's prospective need absent some special defense which would demonstrate their inability to do so. (at 243-244) The issue before the court is whether that obligation should be calculated in strict accordance with the methodology previously adopted in AMG or whether AMG should be modified in any respect. The court will consider each of the township's suggested modifications separately.
The methodology adopted by this court defines present need as the indigenous need of a municipality and the fair share of the reallocated excess need of the municipality's present need region. Indigenous need is defined as substandard housing currently existing in any municipality. Every municipality, regardless of its characterization in the SDGP, is responsible for meeting its own indigenous need. However certain municipalities, even though located in areas characterized as growth in the SDGP, have an indigenous need which far exceeds their fair share. They cannot be expected to provide decent housing for a disproportionate amount of the need. (at 243-244) Therefore, when the total regional housing stock is determined and the percentage of that stock which is substandard is identified, any municipality whose indigenous need, in relationship to its housing stock, is in excess of the regional percentage of substandard housing will have its excess assigned to a reallocation pool. This pool will be distributed to all growth area municipalities
[206 NJSuper Page 169]
within the region which do not have substandard housing in excess of the regional percentage, excluding selected urban aid municipalities as identified in the AMG opinion. AMG, 207 N.J. Super. at 537.
The AMG methodology identifies a housing unit as falling within the substandard category if it has any one of the following characteristics:
1. Overcrowded units -- defined as dwelling units occupied by more than 1.01 persons per room
2. Units lacking complete plumbing facilities for the exclusive use of the occupants
3. Units lacking adequate heating
The number of such units is identified by reference to Census Bureau figures in schedules STF-1 and STF-3 of the 1980 Census. Id. at 471-511. In order to identify the units lacking adequate heating, an additional mathematical computation is performed which need not be repeated here. Id. at 461-463. The total of the unduplicated count for these three categories results in the total number of units which the AMG methodology defines as "substandard". However, to obtain the number of substandard units actually occupied by lower income households, the methodology makes one additional adjustment. A 1978 study by the Tri-State Regional Planning Commission entitled, "People, Dwellings & Neighborhoods," reported that 82% of substandard housing units within the area covered by that study were occupied by low and moderate income households. Therefore, the methodology multiplies the gross number of substandard units by 82% thereby reducing the gross number by 18%. The method of reallocation of the excess present need pool is described in AMG. AMG, at 402, 404-405.
Subsequent to AMG, Judge Stephen Skillman, the Mount Laurel judge for the northern district, adopted a different method of identifying the present need obligation of a municipality. Countryside Properties Inc. v. Ringwood Bor., et al., 205 N.J. Super 291 (Law Div.1984). In that opinion Judge Skillman utilized, in part, the Rutgers Center for Urban Policy Research
[206 NJSuper Page 170]
method for identifying present need. "Mount Laurel II: Challenge and Delivery of Low Cost Housing," (1983) (hereinafter CUPR). The report uses seven housing characteristics or "surrogates" obtained from the 1980 Census data in order to identify whether a housing unit should be considered substandard:
1. Year built: prior to 1940 or 1940 and after
2. Persons per room: more than 1.01 persons per room
3. Access to unit: lack of exclusive access
4. Plumbing facilities: lack of exclusive plumbing
5. Kitchen facilities: lack of complete kitchen facilities
6. Heating facilities: lack of central heating
7. Elevator: lack of elevator in a multistory structure of four stories or more (CUPR at 111)
The report establishes a two-level analysis of deficiency depending on whether the unit was built before 1940. If the unit was built before 1940 it will be considered substandard if it has any one of the other six deficiencies listed above. If the unit was built in 1940 or after the unit is substandard if it has any two of the other six deficiencies listed above. It should be noted that the six deficiencies include two which are essentially identical to those used in the AMG methodology -- overcrowding and lack of exclusive plumbing. The CUPR also utilizes lack of central heating as a surrogate. AMG uses the category of units lacking adequate heating. The difference is that the AMG methodology is more restricted in its count. A unit with a room heater vented by a flue is considered adequate heating. AMG counts as deficient only those units having room heaters with no flue, units heated by fireplaces, stoves or portable room heaters, or units having no heating whatsoever. As noted, the CUPR includes as deficient all units not having central heating.
The CUPR data base regarding deficient housing is the New Jersey public use sample which is a five percent sample of all households in New Jersey taken from census data. All information for that sample is stored on computer tapes. The CUPR contends it is possible to determine by a computer run the number of households occupying deficient housing. It is also possible to
[206 NJSuper Page 171]
identify the size and income levels of households occupying these units, to establish which are lower income and to eliminate any double counting of units having multiple deficiencies.
The data for the public use sample is not available at a municipal level. The smallest area for which the data is available is designated as a subregion. A subregion is either a county or an area having a minimum population of 100,000. The CUPR has divided the State into 52 subregions.
In his review of the comparative merits of the AMG and CUPR methodology, Judge Skillman found that he could not accept the Rutgers report in full because of its failure to count substandard units in accordance with the dictates of Mount Laurel II. Specifically, he concluded that overcrowded units must be treated as a separate category. Therefore, an overcrowded unit must be deemed deficient whether or not any other surrogate of deficiency is present. With that modification, Judge Skillman adopted the CUPR method of identifying substandard units. He found a significant weakness in the AMG methodology in that it assumed 82% of the housing units designated as deficient were occupied by lower income households. As noted, that percentage is derived from a report of the Tri-State Regional Planning Commission issued in 1978. Judge Skillman found that the report contains no explanation of how the 82% figure was established. He concluded that the figure was suspect because it encompassed the entire New York metropolitan area, the source of the data appeared to have been the 1970 Census and the term "inadequate" housing as used in the Tri-State Commission report may have included not only physically deficient and overcrowded units but also units occupied by persons who pay a disproportionate percentage of their income for housing or who commute an excessive distance to work.
Finding that the CUPR methodology did not contain any of these apparent weaknesses, Judge Skillman concluded that the computer runs of the public use sample are a more reliable
[206 NJSuper Page 172]
method of calculating the extent of the present need. He did concede that in the case before him the percentage of substandard housing units occupied by non-lower income households appeared "somewhat surprising" but he provided several possible explanations for this phenomenon. Countryside Properties, supra, at 302. I would only add that I reached the same conclusions with regard to the percentages and I am not at all certain that the explanations which are provided in Countryside Properties can fully alleviate my suspicions. Nonetheless, there is no statistical or empirical basis which could demonstrate that there is any significant error in the approach taken in the Rutgers' report.
Judge Skillman concluded that the CUPR methodology is a far more reliable indicator of the percentage of deficient units occupied by lower income households. He noted, however, one significant disadvantage of the CUPR methodology is that the computer data is only compiled for subregions. Therefore, it was necessary to devise a method of converting the data from the subregional level to the municipal level. Judge Skillman found that although the CUPR method was more reliable than the AMG method for determining the total number of deficient units occupied by lower income persons, the AMG method does provide an appropriate means for converting subregional results to the municipal level. He reasoned that it is possible to estimate, using the AMG methodology, the percentage of subregional present need located in the municipality and then to apply that figure to the total present subregional need as determined by the CUPR methodology in order to arrive at a present indigenous need for the municipality involved. Thus, the municipality's present need obligation would be obtained by identifying the total amount of deficient housing in the municipality and the total amount of deficient housing in the subregion pursuant to the AMG methodology. By dividing the municipal deficient housing by the subregional deficient housing a percentage of deficient housing would be produced for that municipality. That percentage would then be multiplied by
[206 NJSuper Page 173]
the number of subregional deficient units as determined under the CUPR methodology in order to calculate the municipality's obligation.
At the time the AMG opinion was authored, the court had before it a copy of the CUPR report. However, no representative of the CUPR was presented as a witness in the AMG trial. Since that time Dr. Robert W. Burchell of Rutgers University testified before Judge Skillman in the Countryside Properties case and, at the court's request, testified in this case. I have therefore had an opportunity to fully reexamine the present need issue in light of Judge Skillman's opinion and the testimony of Dr. Burchell. While this court has heard testimony concerning the high calibre of work performed by the Tri-State Commission, I conclude that its use of the 82% estimate should give way to the apparently more reliable data generated by the CUPR. In all other respects, however, it is my judgment that the approach adopted in AMG is sound.
As noted in AMG, any reasonable methodology must have as its keystone reliable data, as few assumptions as possible and an internal system of checks and balances. Reliable data refers to the best source available for the information needed and the rejection of any data which may be suspect. The need to make as few assumptions as possible refers to the desirability of avoiding subjectivity and any data which requires excessive mathematical extrapolations. The internal system of checks and balances refers to the effort to include all important concepts while not allowing any one concept to have a disproportionate impact. AMG, 207 N.J. Super. at 452-453. While I continue to have some concern about the surprisingly high percentage of deficient units occupied by persons who are not in the lower income category, I must recognize that the CUPR approach to identifying the percentage of lower income households is the most reliable method yet developed and is preferable to the assumption involved in the utilization of the Tri-State Commission report.
[206 NJSuper Page 174]
At the court's request, the CUPR has done an analysis of the total substandard housing units in the 11 county region representing Franklin's present need region. (That region is the same as the present need region involved in AMG.) The report concludes that if the three surrogates of deficiency utilized by the AMG methodology are properly income qualified through the use of the computer tapes, 64.2% of the units in the 11 county region are occupied by lower income households. The court will utilize that percentage instead of the 82% used in the AMG opinion. Since the percentage is lowered, a smaller number of deficient units exist in the region. Therefore, the regional percentage of substandard units to standard units identified in AMG as 6.4% must necessarily change. The percentage of deficient units to standard units will be lower on a regional basis. Since the pool of present need to be reallocated is determined by the relationship of the regional percentage of deficient housing to regional housing stock as compared to the municipal percentage of deficient housing to municipal housing stock, there will have to be a recalculation of the excess pool. Once that number is arrived at, the factors utilized in the AMG opinion for the purposes of reallocation of the excess present need pool to each municipality can come into play. Therefore, I am directing the master (already appointed in this case) to recalculate the excess pool for the 11 county region so that Franklin Township's share of that pool can be determined pursuant to the AMG methodology.
It should also be noted that the same reasons which justify the use of the percentage of deficient units generated by the computer tapes on a regional basis to produce the municipality's reallocated excess obligation, also justify the use of the percentage on a subregional basis in order to determine the indigenous responsibility of the municipality. Thus, the master shall also determine the subregional percentage of deficient housing for the subregion in which Franklin Township is located within the public use sample for the purposes of identifying the township's indigenous need.
[206 NJSuper Page 175]
As noted above, except for the modification of the AMG methodology concerning the use of the 82% factor, the court again endorses the balance of the AMG approach to identifying and allocating indigenous and excess present need. I reject the balance of the CUPR approach for the following reasons:
1. I question the validity of some of the surrogates of deficiency used by the CUPR. In particular, a substantial argument can be made that the mere fact that a unit lacks an elevator in a multistory structure of four stories or more should not signify deficiency. There are many high-priced and substantial structures throughout the state of four stories or more which no one would consider deficient because of the absence of an elevator. The gentrification of old multistory buildings that is taking place along the Hudson River waterfront is a prime example. Those units are selling at $100,000 and higher.
2. The requirement that a unit be lacking in central heating is also open to dispute. The Building Officials Code of America permits new construction of units without central heating if flues are provided.
3. The use of the year 1940 as a breakoff date to create a two-tiered analysis has not been justified by any data. While it can be assumed that older buildings are more likely to be dilapidated, the selection of the year 1940 is neither explained nor supported. Furthermore, the use of 1940 as a breakoff date can produce the anomaly that a building that was constructed in 1939 could be deemed substandard if it had one other deficiency and yet if it were built in 1940 it could become standard because it had only one surrogate of deficiency.
4. The objections listed above could be met with the contention that the use of the seven surrogates broadens the number of indicia of deficiency and thereby provides a more accurate indication of substandard units than a methodology that recognizes a single surrogate as signaling housing dilapidation. Thus, it could be argued that whatever weakness exists in any of the seven surrogates is offset by the greater possibility of identifying truly deficient units. The problem is that even if the court could be persuaded by that argument, its value is more than counterbalanced by the fact that the CUPR approach cannot generate individual municipal numbers through the use of its seven surrogates. As indicated above, the public use sample produces a subregional number which has to be disaggregated to a municipality through a series of assumptions which utilize figures produced by the three surrogates involved in AMG to reallocate seven surrogates produced by the CUPR method. The AMG approach is able to identify, through the use of census data, the specific number of substandard units for each municipality because AMG limits itself to three factors which are available in an unduplicated count in the census material for each municipality in the State.
5. The cost of obtaining deficient unit data to apply the CUPR method can be significant. Each subregional configuration which must be obtained costs approximately $500. In the event that it is necessary to measure a large region consisting of many subregions the cost can be substantial. In the case of the
[206 NJSuper Page 176]
11 county present need region within which Franklin Township is located, there are 34 subregions.
6. The CUPR data is available only through limited sources. The AMG approach can be utilized by any municipal planner who would be able to generate the numbers without the use of the CUPR computer data.
7. The court's experience in obtaining data from the CUPR demonstrates that there has been a significant delay in receipt of the information. The AMG methodology can be applied in a very short time. In fact, once the tables are established for each county, as they must be in order to determine the municipal number, they are available to the planning community in general.
8. Also of major significance is the fact that there is no presently acceptable method to disaggregate the subregional numbers to the municipality. Dr. Burchell confirmed that fact in testimony before this court. Any method that has been suggested to date involves making assumptions which cannot be proven empirically or with any degree of certainty.
9. As Judge Skillman has pointed out, the CUPR does not identify overcrowding as a separate category of deficiency. I agree with his conclusion that it must be so identified. The AMG methodology does separately count those units.
The AMG method of identifying present need has several advantages:
1. It uses three simple, direct indicia of substandardness. As noted in AMG, few would argue that a unit which is overcrowded or which lacks adequate plumbing or heating is not "substandard" as that word is commonly understood. AMG, 207 N.J. Super. at 420-421.
2. These three categories are individually identified in an unduplicated manner in the census data.
3. The three surrogates can be independently identified for each municipality.
4. The three surrogates are clearly reflective of substandardness. Moreover, planners appearing before this court have testified that if one of the three surrogates is present there is a substantial likelihood that other major deficiencies are also present in the unit.
5. The use of the heating deficiency factor in AMG is more in keeping with a reasonable definition of adequate heating inasmuch as many building codes permit heating through the use of a flue even if central heating units are not utilized.
On balance, I am satisfied that the AMG methodology is still the best yet devised, with the modification of the 82% factor. Despite arguments to the contrary, it is the most conservative estimate of present need once the three surrogates are income qualified. At the court's request, Dr. Burchell has produced a comparison of four approaches used to calculate the total number of low and moderate income units in the State and in the 11 county region. The AMG approach using the 82% assumption
[206 NJSuper Page 177]
need not be discussed since it is now abandoned. The other three methods should be compared. As to the entire State, the AMG three surrogate approach, income qualified through the use of the computer tapes, produces a total of 112,440 low and moderate income units. The CUPR method produces a total of 120,120 units. The CUPR method as modified through the addition of a separate identification of overcrowding, (as required by Judge Skillman in Countryside Properties, supra) produces 147,560 units. On a regional basis for the 11 county region, the AMG approach, income qualified by the CUPR data, identifies 82,440 units. The CUPR approach identifies 94,040 units and the CUPR approach modified to include overcrowding as a separate surrogate identifies 112,560 units.
It may be noted, parenthetically, none of the calculations of present need discussed above include in the definition of substandard units those standard units in which lower income households are paying a disproportionate share of their income for housing. The Supreme Court suggested that not more than 25% of a home owner's income should be spent for housing costs. 92 N.J. at 221, n. 8. Since Mount Laurel II was decided, the figure of 28% of household income for sales and 30% for rentals has been widely accepted by the planning community and mortgage lenders. At the court's request, Dr. Burchell prepared a report concerning the possible inclusion of the financial need category within the present need figures. His conclusions were that statewide there are 76,040 low and moderate income owner-households occupying single family structures in which the monthly ownership cost exceeds 28%. There are 268,560 low and moderate income tenants spending more than 30% of their income for household costs. In the 11 county region there are 37,640 households spending over 28% of their income for ownership of single family homes and 191,800 tenants spending more than 30% of their income for household costs. While the court recognizes that there may be an overlap between the units which are substandard in the
[206 NJSuper Page 178]
manner described by the various methodologies and those units which fall into the financial need category, the foregoing figures should give some indication of the magnitude of the need which is not addressed in any of the methodologies. I continue to believe it is not appropriate to include the financial need category as part of the identification of present need. My reasons for that conclusion are set forth in the AMG opinion. AMG, 207 N.J. Super. at 422-424. However, the exclusion of that category together with the demonstrated conservative nature of the AMG estimate of present need refutes any argument that the AMG approach is grossly overstating the present housing need of this State.
Determination of the Commutershed
In the AMG opinion the court determined that the prospective need region for any municipality shall be a modified commutershed measured in all directions from the functional center of the municipality based on a 30 minute drive time. Id. at 400. The justification for the 30 minute drive time is fully set forth in AMG. Id. at 418 et seq. It should be noted here, however, that the 30 minute region includes all counties touched within a 30 minute commute. A longer time could easily be justified both statistically and by prior case law. Oakwood at Madison, Inc. v. Madison Tp., 72 N.J. 481, 528 (1977) The modified commutershed approach used in AMG was designed to obtain a sufficiently large commutershed area which also coincided with county lines.
The 30 minute drive is measured by speeds depending upon the designation of the road as local and county, state and federal or interstate, Garden State Parkway or New Jersey Turnpike. The AMG opinion defined the functional center in a three-tiered classification. It held that the functional center shall be the generally recognized commercial-residential core of the community or what is typically referred to as the "downtown
[206 NJSuper Page 179]
area". In the absence of a commercial-residential core, the functional center is to be the municipal building. Absent either a recognized commercial-residential core or municipal building, the functional center is to be the major crossroads within the municipality. This case demonstrates that there may be instances in which the definition of the functional center leaves room for some subjective evaluation by the court. In most cases, these subjectivities will not make a difference in the configuration of the prospective need region. In this case, however, it does.
Plaintiffs contend that the functional center of Franklin Township is its municipal complex. They argue that there is no recognized commercial-residential core in the township. The township asserts that there is a commercial-residential core located at or near the intersection of Easton Avenue and John F. Kennedy Boulevard in the northeast quadrant of the town. There is no doubt that if the township is correct, the prospective need region for Franklin Township would include the seven counties of Hunterdon, Mercer, Middlesex, Monmouth, Somerset, Union and Morris. Plaintiffs claim that using the municipal complex as the functional center results in a six county configuration which excludes Morris. At the conclusion of the trial, the court was left with substantial doubt as to the actual location of the functional center and as to whether the commutation time had been accurately measured. As a result, the court appointed an expert to render an independent judgment concerning the location of the functional center and its impact on the prospective need region.
The expert's report concludes that the functional center of the municipality is within the Middlebush portion of the town where Routes 549, 619 and 615 intersect and where the municipal building is located. Based on that conclusion and the measurement made of commutation times to the municipal complex, the expert determined that the municipal building falls just outside the 30 minute travel time to Morris County and therefore, Morris County should not be included within the
[206 NJSuper Page 180]
regional configuration. The expert found that the municipal building is almost, but not quite, reached in a 30 minute commute.
The issue before the court in this case is what adjustment, if any, should be made to the methodology adopted in the AMG case when there is substantial doubt concerning the inclusion or exclusion of a county in the commutershed region because of uncertainty relating to the functional center or the measurement of the commutation time from the functional center. The evidence in this case remains in equipoise.
Whether Morris County is included within the commutershed region of Franklin Township does not have a very significant impact upon its prospective housing need obligation under the AMG methodology. Use of the six county configuration results in a prospective need obligation of 2,114 units. Use of the seven county configuration results in a prospective need obligation of 2,060 units.
Based on the fact that the evidence is in equipoise and that the court still has substantial doubt as to the functional center and given the fact that even if the municipal complex is designated as the functional center, it is barely outside a 30 minute commute and finally acknowledging the closeness of the prospective need numbers whether the region is six or seven counties, a fair resolution of the issue would be to average the prospective need numbers for the six county and seven county configurations. The result is that the prospective need obligation of Franklin Township is 2,087 units if all other aspects of the AMG methodology are utilized. The difference is that the obligation of the town is 27 units more than the seven county region and 27 units less than the six county region.
The court has considered all of the other modifications to the AMG methodology proposed by defendant. Most of the suggested
[206 NJSuper Page 181]
changes were before the court in the AMG case and were expressly rejected in the AMG opinion. Indeed, the principal expert used by defendant in AMG was the same expert offered by Franklin Township in this case. To the extent that he or the other expert called by Franklin asserted any new modifications to the AMG methodology, I find the suggestions to be unsound.
Therefore, the prospective need obligation of Franklin Township is set at 2,087 units. The total fair share shall be obtained by adding the indigenous and reallocated present need units which must be calculated by the master pursuant to this opinion. The parties shall, of course, have an opportunity to be heard with respect to that calculation and the issue of credits for prior compliance claimed by defendant. That latter issue will be determined after the master has reviewed the claimed credits, filed his report and the parties have had an opportunity to respond.