United States v. Starrett City Associates, 660 F. Supp. 668 (E.D.N.Y. 1987)

*669 U.S. Dept. of Justice, Civil Rights Div., Housing and Civil Enforcement Section, Washington, D.C., for plaintiff by William Bradford Reynolds, Asst. Atty. Gen., Joan A. Magagna, Linda F. Thome, Jeremiah Glassman, Katrina Grider.

Paul, Weiss, Rifkind, Wharton & Garrison, New York City, for defendants by Morris B. Abram, Peter Buscemi, Michael G. Carey, Daniel L. Sosland.

NEAHER, District Judge.

The Attorney General, on behalf of the United States (hereinafter "the government"), commenced this action pursuant to Title VIII of the Civil Rights Act of 1968, *670 42 U.S.C. §§ 3601 3631, hereinafter referred to as the "Fair Housing Act".^[1] In brief, the government contends that the defendants, who constructed, own and operate the extensive apartment housing complex in the Borough of Brooklyn, known as "Starrett City", have continuously violated § 3604 of the Fair Housing Act. That section makes it unlawful

"(a) To refuse to ... rent ... or otherwise make unavailable or deny, a dwelling to any person because of race, color, ... or national origin.

(b) To discriminate against any person in the terms, conditions, or privileges of ... rental of a dwelling ... because of race, color ... or national origin.

(c) To make, ... or cause to be made, printed or published any notice, statement or advertisement, with respect to the rental of a dwelling that indicates any preference, limitation, or discrimination based on race, color, ... or national origin, or an intention to make any such preference, limitation or discrimination.

(d) To represent to any person because of race, color, ... or national origin that any dwelling is not available for ... rental when such dwelling is in fact so available."

The parties have joined issue by way of cross-motions for summary judgment accompanied by voluminous documentation. The issue to be decided is whether Starrett City's longstanding policy of limiting the percentage of black and hispanic applicants, who will be accepted as tenants, violates § 3604.

Starrett City was privately constructed in Brooklyn on a site originally planned for a 6,000 unit development of cooperative apartments to be named Twin Pines Village. That project, undertaken by the United Housing Foundation (UHF), was abandoned in 1971 despite a $124,000,000 loan extended by the New York State Housing Finance Agency (HFA). Subsequently, Starrett Housing Corporation agreed to take over the development, the State agreeing that it would be operated as rental housing rather than cooperative apartments.

The change from cooperative to rental status required approval by the New York City Board of Estimate because the city real estate tax abatement granted to Twin Pines Village under the Mitchell-Lama program had to be transferred to Starrett City. According to Robert C. Rosenberg, Senior Vice President of Starrett City, Inc. and General Manager of co-defendant Delmar Management Co., who has managed Starrett City since July 1973, there was substantial community opposition to the transfer, including that of the Brooklyn Borough President and a member of Congress from Brooklyn, who threatened to oppose transfer of the federal contract for the payment of interest subsidies. Opposition centered on the fear that, in light of the neighborhood surrounding the project and past experience with subsidized housing, the conversion to rental apartments would result in Starrett City's becoming an overwhelmingly minority development. Nonetheless, the resolution was adopted by vote of 20 to 2, only the Brooklyn Borough President dissenting, upon the assurance of Starrett City's developer that it was intended to create a racially integrated community.

Privately owned and managed, Starrett City is the largest housing development in the United States assisted by both the federal government, through its Department of Housing and Urban Development (HUD), and the State of New York through its Housing Finance Agency (HFA). In contrast to Starrett City's owners, whose capital contribution totals $19,091,000, HFA has increased its mortgage loan to Starrett to $362,720,000, funds which, of course, are derived ultimately from the taxpayers *671 of the State and the United States. In addition, under Section 236 of the National Housing Act of 1934, as amended, 12 U.S.C. § 1715z-1, HUD makes monthly payments to HFA which reduce from 8½% to 1% the interest rate that Starrett City would otherwise have to pay on its mortgage loan. From April 1975 through March 1986, HUD has paid HFA $211,275,605 in reduction of Starrett City's debt service obligation.

Eligible tenants at Starrett City are also the beneficiaries of a HUD Rental Assistance Program ("RAP") and Section 8 subsidies. Originally limited to one-bedroom apartments housing senior citizens, the program of rent subsidies has since January 1981 been expanded to apply to all of the 5,881 apartments. The maximum income limits for families applying for such benefits are as follows:

Number of           Low             Very Low
 Persons           Income            Income 
    1              $15,250           $10,000
    2               17,400            11,000
    3               19,550            12,800
    4               21,750            14,250
    5               23,100            15,400
    6               24,450            16,550

Under the Section 8 agreements, which run for 15 years, Starrett City must rent at least 30% of the Section 8 units (i.e., at least 1000 units) to Very Low Income Families; viz., those whose income does not exceed 80% of the median income for the area. That contractual obligation is required by the governing statute and regulations. 42 U.S.C. § 1437f(c) (7) (1976) (stricken by Pub.L. 97-35 § 322(e) (2), 95 Stat. 357, 402); 24 C.F.R. § 882.113 (1980). The 1981 amendments eliminated the difference between RAP and Section 8 tenants by providing that rents for new RAP and Section 8 tenants would thereafter be set at 30% of adjusted monthly income, and rents for existing RAP and Section 8 tenants would gradually increase, over a five year period, to 30% of adjusted monthly income. 12 U.S.C. § 1715z-1(f) (1982); 42 U.S.C. § 1437a(a). The current funding limit for Section 8 subsidies at Starrett City is $12.5 million.

The first step in the application process for an apartment at Starrett City is the completion of a Preliminary Information Card (P.I. Card), containing the applicant's name, address, race or national origin, family composition, type of apartment sought, and information about the applicant's employment and income. The applicant mails the P.I. card to Starrett City, where rental office personnel determine whether an applicant is preliminarily eligible, based upon income, family composition and size of apartment requested. When an applicant is determined to be preliminarily eligible, a letter is sent advising him or her of apparent eligibility, and indicating that no apartments are available but that the applicant's name has been placed in the active file, and that he or she will be contacted as soon as Starrett can offer an apartment.

The active file is used as a waiting list and is cross-filed in chronological order by apartment size, income and race or national origin. For example, for one-bedroom apartments and applicants with incomes between $15,000 and $20,000, separate files are maintained for white applicants, black applicants, hispanic applicants, mixed-race applicants and others. Similar files are maintained by racial or national origin groupings for persons having income levels for two and three bedroom apartments.

Applicants are selected from the active files for final processing, which consists of final eligibility determination, credit and income checks and a home visit. A pool of fully processed applicants is maintained, from which applicants are offered apartments as vacancies occur. As the pool of processed applicants shrinks, applicants from the active files are added to the pool. To maintain by apartment a racial balance of approximately 64% white, 22% black, and 8% hispanic, the race or national origin of vacating tenants is noted and applicants of similar race or national origin are selected for the pool.

The consistency of Starrett City's tenant selection process over the years is reflected *672 in the following table, which sets forth the number and percentage of units at Starrett City occupied by each racial and ethnic group for the dates indicated:

   Date  Total   White  (%)    Black (%)   Hispanic (%)  Oriental (%)   Other (%)
  12/75  1578    909 (57.6)   369 (23.4)    265 (16.8)      32 (2.0)      3 (0.2)
   1/76  1660    953 (57.4)   393 (23.7)    276 (16.6)      35 (2.1)      3 (0.2)
   2/76  1912   1147 (60.0)   428 (22.4)    289 (15.1)      45 (2.4)      3 (0.2)
   4/76  2309   1428 (61.8)   511 (22.1)    310 (13.4)      56 (2.4)      4 (0.2)
   5/76  2651   1692 (63.8)   557 (21.0)    334 (12.6)      64 (2.4)      4 (0.2)
   3/77  3798   2532 (66.7)   803 (21.1)    382 (10.1)      42 (1.1)     39 (1.0)
   4/77  3898   2602 (66.8)   822 (21.1)    382 ( 9.8)      48 (1.2)     44 (1.1)
   6/77  4098   2717 (66.3)   856 (20.9)    391 ( 9.5)      73 (1.8)     61 (1.5)
   8/77  4364   2867 (65.7)   895 (20.5)    403 ( 9.2)     107 (2.5)     92 (2.1)
   7/78  4744   3004 (63.3)  1013 (21.4)    389 ( 8.2)     181 (3.8)    157 (3.3)
   8/78  4821   3033 (62.9)  1025 (21.3)    406 ( 8.4)     186 (3.9)    171 (3.5)
   9/78  4895   3069 (62.7)  1030 (21.0)    425 ( 8.7)     195 (4.0)    176 (3.6)
  10/78  4969   3103 (62.4)  1050 (21.1)    438 ( 8.8)     203 (4.1)    175 (3.5)
  11/78  5015   3118 (62.2)  1069 (21.3)    449 ( 9.0)     203 (4.0)    176 (3.5)
  12/78  5061   3159 (62.4)  1065 (21.0)    458 ( 9.0)     202 (4.0)    177 (3.5)
   6/79  5340   3395 (63.6)  1109 (20.8)    457 ( 8.6)     201 (3.8)    178 (3.3)
   8/79  5456   3502 (64.2)  1144 (21.0)    460 ( 8.4)     196 (3.6)    154 (2.8)
   1/80  5518   3521 (63.8)  1154 (20.9)    421 ( 7.6)     228 (4.1)    194 (3.5)
  10/81  5773   3693 (64.0)  1174 (20.3)    453 ( 7.8)     281 (4.9)    172 (3.0)
   2/82  5789   3721 (64.3)  1177 (20.3)    447 ( 7.7)     279 (4.8)    165 (2.9)
   1/84  5785   3743 (64.7)  1206 (20.8)    459 ( 7.9)     259 (4.5)    118 (2.0)

Starrett City's tenant selection practices have always limited blacks and hispanics to far fewer apartments than would be expected if race and national origin were not taken into account in the tenant selection process. Conversely, whites have received substantially more apartments than their application rates project. In the rent-up period, fully completed apartments were kept off the market even though eligible blacks and other minorities were available on the waiting list. Thus, in February 1979, blacks constituted over 47% of the waiting list but occupied fewer than 21% of Starrett City apartments in June 1979, whereas whites constituting less than 24% of those on the waiting list occupied more than 63% of the apartments. As of October 1985, blacks made up 53.7% of the waiting list but, as of January 1984, occupied only 20.8% of Starrett City apartments. Similarly, hispanics, who made up 18% of the waiting list, occupied only 7.9% of the apartments. In contrast, whites, who made up only 21.9% of the waiting list, occupied 64.7% of the apartments.

As a result of defendants' racially controlled tenant selection program, the average waiting time for black or hispanic applicants is significantly longer than the average waiting time of white applicants. Defendants do not deny this; in fact, they admit that minorities continue to wait longer for apartments than do whites as a result of their practice of managed waiting lists. This is illustrated by the following table:

          1 Bedroom  2 Bedroom  3 Bedroom
White       4 mos.       2 mos.      2 mos.
Black      11 mos.      20 mos.     18 mos.
Hispanic    5 mos.       4 mos.    9½ mos.

Defendants assert that it is necessary to take race and national origin into account in selecting tenants in order to preserve stable integration at Starrett City. They contend that the discontinuance of their current rental practices, in view of the large number of minority applicants on the waiting list and the proportionately higher moveout rate of whites, would result in a predominantly minority tenant population at Starrett City. In fact, they hope to *673 achieve a substantially larger proportion of white applicants as the result of a newly adopted advertising campaign for Starrett City and additional residential developments in the area. Defendants predict that it will be at least fifteen years before they will stop selecting tenants on the basis of race and national origin.^[2]

In contrast to their current position is a document defendants signed December 11, 1973, which was approved by the State Commissioner of Housing and Community Renewal, and which specifically provided:

"The Owner and the Agent [Starrett City, Inc. and Delmar Management Co.] will not discriminate against any applicant for an apartment or any tenants already in occupancy because of sex, race, creed, color or national origin. The Owner and the Agent will take affirmative action to insure open occupancy and racial integration in the operation of the project. Such action shall be taken with reference, but not to be limited to: advertising, use of available subsidies such as capital grant funds, H.O. P.E. loans, interest payment subsidies and other aids which may be used to encourage integrated occupancy of the project." (Emphasis supplied).

Defts.Exh. G, Notice of Cross-Motion for Summary Judgment. That agreement was reapproved when the Starrett parties later amended it. Moreover, when defendants' Affirmative Marketing Plan was approved in June 1975, they were made aware that "different processing procedures" could not "be utilized on the basis of race", in dealing with minority or majority persons in furthering their marketing plan. Defts. Exh. D.

And again on January 19, 1981, when Starrett City, Inc. entered into an Amendatory Rental Assistance Contract with HUD, which increased the amount of payments to $9,389,543 for 2,538 rental assistance units, it was reminded of the following "NONDISCRIMINATION" clause:

"(a) General. The Owner shall not in the selection of Families, in the provisions of services, or in any manner, discriminate against any person on the grounds of race, color, creed, religion, sex, national origin, or handicap.

(c) * * * The owner shall comply with all requirements imposed by Title VIII of the Civil Rights Act of 1968, which prohibits discrimination in the * * * rental, * * * of housing on the basis of race, color, * * * or national origin * * *."

See Govt's Attachment 5 in Opposition to Defendants' Motion to Dismiss, pp. 13-14.

Defendants maintain that their race-conscious practices at Starrett City were adopted at the behest of the State solely to achieve and maintain integration and were not motivated by racial animus. Crediting fully defendants' stated intentions, it is clear that their treatment of blacks and other minority persons constituted a clear violation of the Fair Housing Act. As previously noted, Starrett City tenant selection practices have always limited blacks and hispanics to far fewer apartments than would be expected if race and national origin were not taken into account in the tenant selection process.

In further support of their rental practices, defendants have submitted the written testimony of three experts in the housing field. The Court has studied their testimony and opinions carefully. Oscar Newman, an architect and city planner, in brief summary stated that Census Bureau figures in 1980 showed blacks forming 11.7% of the population, 75.9% living inside urbanized areas, of which 75.4% live in central cities, in contrast to only 42.1% of whites. Between 1960 and 1982 the number of black families with children under 18 increased from 20.7% to 46.5%. The comparable figure for white families was 15.0% in 1982. The average income for all black families *674 with children is 56% of the average for white families. However, with two working partners, income increased to 84% of that of white families with two working parents. In 1978, 52.4% of blacks had completed less than 12th grade, compared with 32.1% for whites. The unemployment rate for blacks is significantly higher than for whites. Among black teenagers it hovers about 50%.^[3]

Newman has observed that as communities become both poor and predominantly black, gradual abandonment occurs. Examining the Starrett City area, he found that neighborhood residential areas that had become more than 50% black also suffered a serious loss in total population. Thus, East New York, 70% black in 1970, lost 33,000 (36%) by 1980, and Brownsville, 72% black in 1970, lost 25,000 (35.5%).

The loss of white residents resulting in the transition of a community to a predominantly black population is commonly referred to as "tipping". Newman acknowledges that the "tipping point is a quantity that is difficult to predict with precision. It has been variously estimated, in different factual contexts, as ranging from a low of 1% black to a high of 60% black." He adds, however, that "Most social scientists and housing experts agree that under normal circumstances tipping begins to occur at between 10% and 20% black occupancy." Still, he acknowledges that "the tipping point for a particular housing development, depending as it does on numerous factors and the uncertainties of human behavior, is difficult to predict with precision." As the Court has already noted, the tipping "uncertainties" did not deter defendants from increasing the minority ratio to 35% over a 5-year period in order to relieve themselves of the Arthur litigation.

Dr. Kenneth Clark, a social psychologist and author of more than 130 works dealing with equal rights and the problems confronting blacks, was retained in 1979 by the New York State Division of Housing and Community Renewal (DHCR) to conduct a study concerning integration at Starrett City. Based upon his research for the State and his survey of "tenant perceptions", he concluded, in an affidavit on behalf of defendants, that the percentage of minorities which could trigger tipping "is probably around 40% on a population basis".

His report, to the State issued in January 1980, entitled Starrett City: A study of Ways of Maintaining an Integrated Residential Community, is Exhibit C to his affidavit. Referring to the New York State Human Rights Law, N.Y.Exec.Law § 292 subd. 19 (McKinney's 1982), which states that "the term `discrimination' shall include segregation and separation", Dr. Clark advocates that it is "imperative to define discrimination in terms of preventing segregation and separation." He recognizes, however, that federal law as well as state and local housing legislation prohibits discrimination on the basis of race, and that none give any affirmative mandate to integrate.

According to Dr. Clark, "[t]he primary aim of the Starrett City Plan was to attract white ethnic tenants" as stated in the 70/30 goal appearing in the marketing plan. He also notes "that after filling the initial vacancies, a waiting list in chronological order by date of application would be kept." If by that statement, it was Dr. Clark's understanding that it was Starrett's future intention to fill vacated apartments on an open occupancy basis, it was certainly not followed. As heretofore noted, blacks and hispanics have always been limited to far fewer available apartments than would be expected, if race and national origin were not taken into account. As Dr. Clark noted, "[t]he goal of [HUD marketing regulations] is to seek means of obtaining and *675 maintaining an integrated community without resorting to arbitrary discrimination." The 30% limitation placed on black and other minority applicants appears all the more arbitrary when, as Dr. Clark's study for the State reveals that non-whites at Starrett City have the same educational level as whites (just beyond high school), and the median earnings of the minority respondents are appreciably higher, viz., $15,294 as compared with $13,400 for the white tenants. See Exh. C to Clark affidavit at 29.

Furthermore, the same sample of white and minority residents established that Brooklyn was the place of previous residence for the greater proportion of minority (75.6%) and white (69.8%) Starrett City residents. Moreover, two out of every three (66%) received rent subsidies: two of every five whites (40%) and one of four minority (26%). In the 40-64 age group, one in five (19%) of the whites and one in twelve (8.3%) of the minorities were subsidized. In the over-65 group, 66.6% of the whites and 60% of the minorities received subsidies.

Moreover, the same sample suggested that minority tenants resided longer at Starrett City than whites. The average number of years for all tenants was 2.58: for whites, 2.0 years and for minorities 2.6 years. One-third (31.8%) of the whites in the 25-39 age group had lived there one year or less. In the over-65 group, 41% of whites and 60% of minorities had lived there more than three years, and none less than a year.

The sample also reflected that nearly half of the white respondents (46%) were not put on a waiting list as prospective renters, whereas only one in four of minority applicants were not. One in five of the minorities (18.9%) waited one year and one in 9 (10.8%) waited two years or more before being placed. In contrast, one in 25 of whites (3.5%) waited one year, none waited two years, and less than 50 (1.6%) waited two years or more.

A third study of a sample of 300 black representatives of households in Starrett City by Dr. Clark's organization (Exh. D to his affidavit) focused on 1) their views about residential integration, 2) their perceptions and expectations of racial integration at Starrett City, 3) the relation between their perceptions and expectations of housing integration and their decisions to remain or move, and 4) their views as to the options management might exercise to achieve and maintain racial integration at Starrett City. A key finding of the study indicated that the great preponderance of black residents strongly affirm residential integration, four out of five stating it was important to their decision to move to Starrett City. However, when it came to the extent of integration, the proportion of blacks and other minorities most frequently cited was 40 to 50 percent. Similarly, when asked "If you knew that the waiting list was predominantly black, would you favor management accepting new tenants ... without regard to race?", a nonrestrictive policy was favored by 58.3%.

The foregoing sample of black residents were predominantly in the prime age category. The median age was slightly over 45, one in five was between 50 and 65, and one in four was 65 and over. Further, the sample was heavily weighted with females, quite well educated persons and white collar workers, placing them in the middle class (45.5%), upper middle (6.9%) and lower middle (8.3%). The sample was evenly divided between those who were married and those in the single, divorced, separated and widowed categories. Regarding level of education, 59.3% held a high school diploma and some college (including trade school) and 12.3% held a four year college degree or more. Regarding income, 37.3% were in the under $15,000 category, 36.3% were in the $15,000 to $25,000 category, and 11.1% were in the $25,000 plus category. Other tenants (15.3%) declined to reveal their income. Regarding occupation, 27.7% were in the professional, managerial or proprietor class; 7.3% were in clerical or sales; 4.7% were skilled craftsmen; 4.0% were unskilled operative; 5.7% were service workers; 4.0% were in City service; and 39.0% were in the homemaker, retired, unemployed or student categories.

*676 The third expert relied on by defendants, Morris Milgram, has been involved in integration efforts since the 1950's. His early efforts to establish small group housing developments in Philadelphia, Princeton, N.J., and Delaware convinced him that a 55%/45% goal did not satisfy whites, who soon left the Philadelphia development, which eventually became 96% black. However, in another development in all-white northeast Philadelphia, consisting of 19 single-family homes, a 2/3 majority- 1/3 minority proved successful. Again, in 1962-64, his group purchased high-rise apartment buildings in Washington, D.C. and neighboring Silver Spring, all of which were integrated without incident. These acquisitions were made, however, prior to the enactment of the Fair Housing Act. Moreover, the 30% ceiling has already been breached by defendants in order to settle the Arthur lawsuit; and Milgram himself has written that "large-scale racial integration in housing requires (1) total open occupancy allowing minorities complete access to all housing ..." Milgram, Good Neighborhood: The Challenge of Open Housing (New York, Norton, 1977: 208).

The government contends that defendants have continually violated sections 804(a), (b), (c) and (d) of the Fair Housing Act, 42 U.S.C. § 3604(a), (b), (c) and (d), by their consistent practice of making unavailable to blacks and hispanics rental apartments which are then made available to whites; and by conditioning rental to minorities on the basis of a "tipping" formula that depends solely on the race or national origin of other tenants at Starrett City. As a consequence, minority applicants must wait up to ten times as long as the average white applicant for an apartment. In addition, employees of defendants are instructed to rent apartments by favoring white applicants in order to maintain a racial balance among tenants. Moreover, defendants send a letter to each applicant placed on the waiting list stating that "at this time an apartment is not available for you", which remains operative until the applicant is invited to come in to complete the processing for an apartment. To the extent that apartments are in fact available but are being reserved for white applicants, the representation of unavailability to qualified black and hispanic applicants is prohibited conduct condemned by the statute.

The government further contends that even if HUD had approved defendants' Affirmative Fair Housing Marketing Plan, that would not permit "the use of managed waiting lists to achieve integration." Nor would the fact that HUD representatives made "numerous visits to Starrett City ..." and "knew that integration could not be maintained ... without the use of race as a factor in the renting of apartments." Indeed, as the government points out, this Court has previously noted that HUD "[s]trongly disavowed ... Starrett's apparent assumption that HUD supported racially based tenant selection practices at Starrett City." See United States v. Starrett City Associates, 605 F. Supp. 262, 264 n. 3 (E.D.N.Y.1985).

Furthermore, as the government points out, even assuming that HUD officials had given approval to defendants' discriminatory practices, defendants would not be insulated from the consequences of their acts. Although the Supreme Court recently spoke to this issue in a different context, Heckler v. Community Health Services of Crawford County, 467 U.S. 51, 104 S. Ct. 2218, 81 L. Ed. 2d 42 (1984), declining to adopt a flat rule that the government may never be estopped under any circumstances, it acknowledged that strong arguments could be made in support of such a rule, stating,

"When the Government is unable to enforce the law because the conduct of its agents has given rise to an estoppel, the interest of the citizenry as a whole in obedience to the rule of law is undermined. It is for this reason that it is well settled that the Government may not be estopped on the same terms as any other litigant."

Id. at 60, 104 S. Ct. at 2224. This Circuit has similarly held that equitable estoppel is available against the government only "in the most serious of circumstances." United *677 States v. Repass, 688 F.2d 154, 158 (2d Cir. 1982). And more recently a district court in this Circuit held that "HUD has no power to excuse discriminatory acts or to waive, on behalf of those injured by them, the right to seek a remedy." United States v. Yonkers Bd. of Education, 624 F. Supp. 1276, 1375 (S.D.N.Y.1985).

The Fair Housing Act requires no critical exegesis to make its meaning clear. In plain language, it is a "prohibition against discrimination in the sale or rental of housing", with exceptions not here relevant. Affirmatively, its purpose was, as stated by its principal sponsor Senator Mondale, the "Outlawing [of] discrimination in the sale or rental of housing ..." and the gradual prohibition of "discrimination on account of race, color, religion or national origin in the sale or rental of housing." 114 Cong.Rec. 2272 (1968). As he further pointed out, the source of that prohibition is the Fourteenth Amendment, which in its Equal Protection Clause forbids a State to deny any person within its jurisdiction the equal protection of the laws. Id. at 2273.

Although the Fair Housing Act does not in terms so state, the Supreme Court "has expressly recognized that substantial benefits flow to both whites and blacks from interracial association and that Congress has made a strong national commitment to promote integrated housing." Linmark Associates, Inc. v. Township of Willingboro, 431 U.S. 85, 94-95, 97 S. Ct. 1614, 1619, 52 L. Ed. 2d 155 (1977) (citing Trafficante v. Metropolitan Life Ins. Co., 409 U.S. 205, 93 S. Ct. 364, 34 L. Ed. 2d 415 (1972)). But Trafficante was brought to restrain such unlawful practices as "manipulating the waiting list for apartments, delaying action on ... applications, using discriminatory acceptance standards and the like." 409 U.S. at 208, 93 S. Ct. at 366, which prompted the government to bring this action. Thus, the gloss placed upon the Act by its principal sponsor regarding the importance of integrated living, and the Supreme Court's approval of interracial association, cannot lawfully be interpreted to mean that where a large housing complex such as Starrett City is concerned, black and other minority applicants who are qualified cannot obtain available apartments on an equal basis with whites but must remain subject to the landlord's quota.

The use of a rigid quota, however, as employed by defendants is not lawful, as was made clear in Williamsburg Fair Housing Committee v. New York City Housing Authority, 493 F. Supp. 1225 (S.D. N.Y.1980). In that case, the Court granted a permanent injunction barring project managers from using a rigid quota which assigned 75% of project apartments to whites, 20% to hispanics and 5% to blacks. As the Court pointed out,

"Such a non-White family would, at least for a time, have been denied an apartment for which it was eligible. Stated differently, that family would have been discriminated against in the privileges of a rental. Title VIII requires no more to establish a violation. 42 U.S.C. § 3604(b)."

Id. at 1248.

The Starrett defendants rely heavily on Otero v. New York City Housing Authority, 484 F.2d 1122 (2d Cir.1973), which arose from a controversy over the allocation of apartments in newly constructed housing in the lower East Side of Manhattan, built for the New York City Housing Authority. The Otero plaintiffs, former residents of razed housing in that area, had been given first priority for apartments in the new housing pursuant to the Authority's regulations. Controversy arose when the Otero plaintiffs learned they were being allotted less than half of the 360 available apartments in order to provide housing for non-former site occupants. The district court concluded, in view of the Authority's affidavit that the surrounding area population was approximately 48.3% white and 51.7% non-white, that the situation was "not one in which an influx of non-whites may provoke the `tipping' of a previously integrated neighborhood, since by definition former site occupants are persons who originally *678 resided in the neighborhood ..." Id. at 1128-30. Accordingly, the district court granted summary judgment in favor of the plaintiffs.

On the Authority's appeal, the Court recognized that "the community is presently integrated, with a racial balance that is almost equally divided between white and non-white residents, but the housing authority seeks to stem a steady decline in the percentage of the white population in the community." Id. at 1124. Nonetheless, the Court observed that "the Authority's denial of housing to a family because of its race could, whether or not labelled a `benign' quota, constitute a form of unlawful racial discrimination in violation of the family's constitutional rights. For these reasons the burden on the Authority is a heavy one." Id. at 1136.

The appellate Court's focus on "constitutional rights" in Otero is of no assistance to defendants. The Fair Housing Act, undergirded by the Fourteenth Amendment, is the basis of the government's case here, and the source of the rights of black and other minority applicants to be treated on the same basis as whites in seeking available housing at Starrett City. The defendants, as private landlords, are not clothed with governmental authority. Their obligation is simply and solely to comply with the Fair Housing Act.

The Court has minutely reviewed the extensive submissions of the parties and has concluded that the government's challenge to Starrett City's tenanting procedure is correct. Defendants cannot arrogate to themselves the powers of the New York City Housing Authority described in the Otero case, which could grant or deny housing subject only to State or Federal law, as in Williamsburg Fair Housing Committee, supra. The applicable law in this case is clearly Title VIII, 42 U.S.C. § 3604, and not the constitutional cases discussed in Otero and relied on by defendants.

Defendants were not authorized to treat the so-called "tipping" phenomenon as a basis for denying minority applicants the same rights white applicants enjoyed with respect to obtaining apartments at Starrett City. Their own expert, Oscar Newman, acknowledged the wide elasticity of that standard, which ranged "from a low of 1% black to a high of 60% black." Indeed, defendants had no difficulty stretching their quota to 35% when it became necessary to avoid litigating the private Arthur lawsuit which threatened their unlawful rental practices.

Defendants' disregard of the requirements of the Fair Housing Act is all the more difficult to understand in light of the findings of their other expert, Dr. Kenneth Clark, based on his survey of black tenants living in Starrett City. They reveal that Brooklyn was the place of residence of the greater proportion of both minority and white tenants. Even more revealing was the finding that white tenants are more subsidized than minority tenants; that the black tenants included substantial numbers of high school and college educated persons; that about a third of the sample survey indicated that their earnings were under $15,000 a year, another third were in the $15,000 to $25,000 category, and 11% were in the $25,000 plus category. Another 15% declined to reveal their income. Regarding occupation, 27.7% were in the professional, managerial or proprietor class; 7.3% were in clerical sales; 4.7% were skilled craftsmen; 4.0% were unskilled operative; 5.7% were service workers; 4.0% were in City service; and 39% were in the homemaker, retired, unemployed or student categories.

Defendants, as private landlords, were not empowered to establish quotas limiting the number of apartments to be made available to eligible minority applicants, in disregard of the Fair Housing Act. Nor were they entitled to hold off the market vacant apartments, which could have been rented to eligible minority applicants, solely to insure occupancy by white persons. Defendants admit that they treat black and hispanic applicants differently *679 than white applicants on the basis of their race and national origin. Such differential treatment violates the Fair Housing Act whatever its motivation.

Accordingly, it is hereby ORDERED, ADJUDGED AND DECREED that the plaintiff, United States of America, have judgment against the defendants as follows:

(1) Permanently enjoining defendants from discriminating against black, hispanic and other minorities on the basis of race, color, or national origin in connection with their applications for available apartments at Starrett City.

(2) Requiring defendants to inform their agents and employees of their policy of nondiscrimination and of their duties under this order and the Act.

(3) Requiring defendants to take steps to notify black, hispanic and other minority applicants presently on the waiting list, future applicants, and the public, of their policy of nondiscrimination.

(4) Requiring defendants to adopt written, objective, uniform, nondiscriminatory tenant selection standards and procedures to be approved by the United States and this Court.

(5) Requiring defendants to maintain recordkeeping and reporting to enable the United States to monitor compliance with this order.

(6) Jurisdiction shall be retained in this Court for a period of three years.

The Clerk is directed to enter judgment accordingly this 5th day of May, 1987.

SO ORDERED.

[1] Pursuant to § 813 of the Civil Rights Act of 1968, the Attorney General may bring a civil action in any appropriate United States district court when he has reasonable cause to believe "that any person or group of persons is engaged in a pattern or practice of resistance to the full enjoyment of any of the rights granted" by the Act. Trafficante v. Metropolitan Life Ins. Co., 409 U.S. 205, 209, 93 S. Ct. 364, 366, 34 L. Ed. 2d 415 (1972).

[2] However, in March 1985, in order to terminate an earlier action brought by black citizens complaining of defendants rental practices, defendants agreed to increase the apartments available to black and hispanic applicants by 175 units at the rate of 35 units per year over a five-year period, subject to the availability of vacancies. Arthur v. Starrett City Associates, 79 CV 3096 (E.D.N.Y.1984) (settlement decree).

[3] According to a New York City housing survey in 1984, the problem of decent housing for blacks and other minorities was complicated by the fact that 57,000 rental units were converted to cooperative housing, which limited rental opportunities for blacks and minorities. A similar situation was noted in Brooklyn, where the house count remained unchanged but reflected a decline of 9700 renters and an increase of 11,900 new homeowners. In the same period the incomes of black renters increased by 20%, which brought the median to 63%, as high as whites. Govt.Exh. 37.