Shelter Corp. v. Ontario (Human Rights Comm.) (No. 2), 1998 CanLII 29852

Shelter Corp. v. Ontario (Human Rights Comm.) (No. 2)

1998-12-22

Ontario Board of Inquiry

Dawn Kearney, J.L. and Catarina Luis

Complainants

and

Ontario Human Rights Commission

Commission

v.

Bramalea Limited (now Bramalea Inc.), The Shelter Corporation and Creccal Investments Ltd.

Respondents

Date of Complaints:

September 23, 1988, June 25, 1990 and May 4, 1992

Date of Decision:

December 22, 1998

Before:

Ontario Board of Inquiry, Deborah Leighton, Ajit John and Mary-Woo Sims

Decision No.:

98-021

Appearances by:

Anthony D. Griffin, Counsel for the Commission

Beth Symes and Bruce Porter and Catherine Bruce (Student-at-law), Counsel for the Complainants

Ian Scott, Christopher Dassios, Stephen Goudge and Martine Doane, Counsel for the Respondents Bramalea Inc. and The Shelter Corp.

Gary Luftspring, Counsel for the Respondent Creccal Investments Ltd.

Thomas Wall, for the Intervener City of Toronto, Non-Profit Housing Corp.

Nichola Savin, for the Interveners Jessie's Centre for Teenagers, Young Mothers' Resource Group and Housing Equality for Youth

Anthea Pascaris, for the Interveners Federation of Metropolitan Toronto Tenants' Associations and United Tenants of Ontario

Allan Maclure, for the Interveners Children's Aid Society of Metropolitan Toronto, Catholic Children's Aid Society of Metropolitan Toronto, Pape Adolescent Resource Centre, Native Child and Family Services and Jewish Family and Child Service

Melinda Rees, for the Interveners National Anti-Poverty Organization, the Charter Committee on Poverty Issues and the Ontario Coalition Against Poverty

Jack Martin, for the Interveners New Experiences of Latin American Refugee Women, Canadian African Newcomer Centre of Toronto, Toronto Refugee Affairs Council and Centre for Spanish Speaking People

Bruce Feldthusen, for the Interveners The Supportive Housing Coalition of Metropolitan Toronto, Robin Gardner Voce Non-Profit Homes Inc., Woodgreen Community Housing Inc. and Housinfo Housing Development Resource Centre of Ontario

Sheena Scott, for the Interveners Coalition of Youth Serving Agencies — Covenant House and Touchstone Youth Centre, Stop 86

Katryn Pereira, for the Interveners Disabled Women's Network Ontario and The Coalition of Visible Minority Women

SYSTEMIC DISCRIMINATION — pattern of conduct discriminatory on the basis of income — HOUSING ACCOMMODATION — rental accommodation denied on the basis of income criteria — SOURCE OF INCOME — housing accommodation denied to recipient of social assistance benefits

DISCRIMINATION — adverse effect discrimination — direct discrimination — bona fide justification — definition of bona fide justification — Etobicoke test (non-employment — BUSINESS NECESSITY — housing accommodation policy discriminatory for economic reasons — REASONABLE ACCOMMODATION — duty to accommodate short of undue hardship — definition of undue hardship — REMEDIES — prohibition on using income criteria for housing accommodation

PRODUCTION OF DOCUMENTS — adequacy of disclosure prior to hearing — duty to disclose — PROCEDURE — adjournment pending production of documents — EVIDENCE — disclosure of evidence

BURDEN OF PROOF — elements of a prima facie case — PARTIES — intervener status for friend of the court (amicus curiae)

Summary: The Board of Inquiry finds that the respondents' use of income criteria or a rent-to-income ratio to exclude three complainants from housing discriminates, contrary to ss. 2(1), 4, 9 and 11 of the Ontario Human Rights Code, on the grounds of sex, marital status, citizenship, place of origin, family status, receipt of public assistance and age. The Board finds also that the application of income criteria was not shown by the respondent landlords to be reasonable or bona fide. Nor did the respondents show that not being permitted to use income criteria to screen tenants would constitute an undue hardship.

In this case the evidence showed that each of the complainants had been denied housing because of the application of income criteria. The complainant Ms. Luis alleged that the use of income criteria constituted discrimination on the basis of sex, marital status, citizenship, place of origin, family status, receipt of public assistance and race. Ms. Luis was a single, black woman, and a refugee from Angola, in receipt of family benefits assistance. She had a young daughter. She alleged that her application for an apartment at The Crossways was denied by the respondent Creccal because of the application of a monthly income requirement of $2,000 which she could not meet.

J.L. was 16 years old and was seeking accommodation for herself and a friend who was 18 years old. J.L. alleged that their application for an apartment at Aquitaine Shores was denied by the respondent Shelter Corporation because of the application of a minimum yearly income requirement of $32,000 which the landlord stipulated applied to each of the two tenants, even though for calculating the income of couples the landlord accepted combined income. They could not meet the requirement. J.L. alleged discrimination on the basis of age, sex and marital status.

Ms. Kearney, a 17-year-old woman who was pregnant and seeking accommodation for herself and her husband who was 18 years old, alleged that use of income criteria by the respondent Bramalea constituted discrimination on the basis of age and sex. She was told that she and her husband would need to have an income of $30,000 per annum to qualify as renters.

The Commission's position was that there was adverse effect discrimination. The complainants characterized the discrimination as systemic discrimination through the use of rent-to-income ratios, that results in direct discrimination against some protected groups.

The respondents did not dispute that they used income criteria based on rent-to-income ratios of 25 percent, 27 percent, 30 percent, 33.33 percent and 34 percent. However, the respondents maintained that the Code does not prohibit credit granters such as landlords from determining who they should extend credit to on the basis of ability to pay. They argued that the use of income criteria does not constitute discrimination. The respondents Shelter Corporation and Bramalea argued that the use of income criteria is bona fide and permitted by the Code under s. 11. The respondent Creccal disputed that it had actually denied an apartment to Kearney since it had no apartments available.

The Board considered extensive expert evidence. The Board finds that income criteria differentially affect equality-seeking groups, defined on the basis of sex, marital status, family status, age, citizenship, race, immigration status, place of origin and being in receipt of income assistance. Women, single mothers, refugees, immigrants, and people receiving social assistance figure disproportionately among those with lower incomes who are excluded from accommodation by the application of income criteria.

The Board finds that young people are hardest hit by income criteria and that young unattached women are virtually excluded from accommodation when income criteria are applied.

The complainants were denied housing based on income criteria derived from rent-to-income ratios. The Board finds rent-to-income ratios presently used by landlords are not valid criteria for assessing a person's ability or willingness to pay rent, nor do they accurately predict a person's likelihood of defaulting on rent payments.

The Board finds that the respondents presented no evidence that the use of income criteria is a reliable predictor of either default or ability to pay. Further, the respondents did not tender evidence to show that being precluded from using income criteria would cause undue hardship.

This leads the Board to conclude that the respondents have not shown that failing to meet the income criteria is a bona fide justification for refusal to rent. The practice of using rent-to-income ratios is not shown to have a connection to the reduction of risk of default and therefore is not reasonable. Neither the subject nor objective element of the defence is established. Nor was undue hardship established.

The Board finds that Ms. Luis was discriminated against on the basis of all of the grounds cited in the complaint. It also finds that J.L. and Kearney were discriminated against on the basis of age and sex.

The Board declines to determine whether there is direct discrimination as alleged. The Board finds that there is adverse effect discrimination or constructive discrimination, and that the appropriate remedy is to strike the discriminatory rule because its impact is pervasive. Citing Colfer v. Ottawa Police Commission and Griggs v. Duke Power, the Board finds that it is proper in cases where the impact of a rule is so pervasive to strike down the rule in its general application.

The Board declares that the use of rent-to-income ratios and minimum income criteria violate ss. 2(1), 4, 9 and 11 of the Code, whether used alone or in conjunction with other selection criteria or requirements. It orders the respondents Creccal Investments, Shelter Corporation and Bramalea to cease and desist from using rent-to-income ratios or minimum income criteria in selecting prospective tenants, whether alone or in conjunction with other selection criteria.

It also orders the respondent Creccal to pay Ms. Luis specific damages of $460 for losses arising when she was denied housing at The Crossways, and) $5,000 as general damages for the breach of ss. 2(1), 9 and 11 of the Code, with pre-judgment interest on these sums.

The Board orders Shelter Corporation to pay J.L. $4,000 as general damages for its breach of ss. 2(1), 4, 9 and 11 of the Code, with pre-judgment interest on that amount.

The Board orders Bramalea to pay Ms. Kearney $4,000 as general damages for its breach of ss. 2(1), 4, 9 and 11 of the Code, and pre-judgment on interest on that amount.

The Board remains seized of the case should there be any difficulties in the implementation of the orders.

[Ed. Note: See also (No. 1) (1994), C.H.R.R. NP/94-43 (Ont. Bd.Inq.).]

CASES CITED

Canadian National Railway Co. v. Canada (Human Rights Comm.) and Action travail des femmes, 1987 CanLII 109 (SCC), [1987] 1 S.C.R. 1114, 8 C.H.R.R. D/4210: 104

Central Alberta Dairy Pool v. Alberta (Human Rights Comm.), 1990 CanLII 76 (SCC), [1990] 2 S.C.R. 489, 72 D.L.R. (4th) 417, 12 C.H.R.R. D/417: 179

Central Okanagan School Dist. No. 23 v. Renaud, 1992 CanLII 81 (SCC), [1992] 2 S.C.R. 970, 95 D.L.R. (4th) 577, 16 C.H.R.R. D/425: 171

Colfer v. Ottawa Police Comm. (January 12, 1979), (Ont. Bd.Inq.) [unreported]: 182

Gibbs v. Battlefords and Dist. Co-operative Ltd., 1996 CanLII 187 (SCC), [1996] 3 S.C.R. 566, 27 C.H.R.R. D/87: 103

Griggs v. Duke Power Co., 401 U.S. 424 (4th Cir. 1971): 143, 182, 194

Janssen v. Ontario (Milk Marketing Board) (1990), 1990 CanLII 12505 (ON HRT), 13 C.H.R.R. D/397 (Ont. Bd.Inq.): 107, 172

Large v. Stratford (City), 1995 CanLII 73 (SCC), [1995] 3 S.C.R. 733, 24 C.H.R.R. D/1: 166

Ontario (Human Rights Comm.) v. Etobicoke (Borough), 1982 CanLII 15 (SCC), [1982] 1 S.C.R. 202, 3 C.H.R.R. D/781: 105, 166

Ontario (Human Rights Comm.) v. Ontario (Human Rights Board of Inquiry) (1993), 1994 CanLII 10981 (ON CTGDDC), 115 D.L.R. (4th) 279, 1993 CanLII 16421 (ON CTGDDC), 20 C.H.R.R. D/498 (Ont. Ct. (Gen.Div.)): 11, 17

Ontario (Human Rights Comm.) and O'Malley v. Simpsons-Sears Ltd., 1985 CanLII 18 (SCC), [1985] 2 S.C.R. 536, 7 C.H.R.R. D/3102: 105, 142, 182

Quesnel v. London Educational Health Centre (1995), 1995 CanLII 18159 (ON HRT), 28 C.H.R.R. D/474 (Ont. Bd.Inq.): 144

R. v. Stinchcombe, 1991 CanLII 45 (SCC), [1991] 3 S.C.R. 326: 11

Sehdev v. Bayview Glen Junior Schools Ltd. (1988), 1988 CanLII 8934 (ON HRT), 9 C.H.R.R. D/4881 (Ont. Bd.Inq.): 107

Thorne v. Emerson Electric Canada Ltd. (1993), 1993 CanLII 16439 (ON HRT), 18 C.H.R.R. D/510 (Ont. Bd.Inq.): 106, 172

Walbar Machine Products of Canada Ltd. v. Ontario (Human Rights Comm.) (1980), 1980 CanLII 3911 (ON HRT), 1 C.H.R.R. D/228 (Ont. Bd.Inq.): 12

Zurich Insurance Co. v. Ontario (Human Rights Comm.), 1992 CanLII 67 (SCC), [1992] 2 S.C.R. 321, 16 C.H.R.R. D/255: 103, 140, 189

LEGISLATION CITED

Ontario

Courts of Justice Act, R.S.O. 1990, c. C.43, s. 27: 198

Family Benefits Act, R.S.O. 1990, c. F.2: 44

General Welfare Assistance Act, R.S.O. 1990, c. G.6: 44

Human Rights Code, R.S.O. 1990, c. H.19

preamble: 193

s. 2: 197

s. 2(1): 198

s. 4: 197

s. 9: 197

s. 10: 193

s. 11: 106, 141, 171, 182, 198

s. 11(1)(a): 145, 165

s. 24(1)(b): 166

s. 35(1): 1

s. 38(1): 1

s. 39: 15

s. 39(4): 15

s. 41: 196

Statutory Powers Procedure Act, R.S.O. 1990, c. S.22

s. 8: 12, 18

s. 12: 16

s. 12(1)(b): 18

AUTHORITIES CITED

Keene, Judith, Human Rights in Ontario, 2d ed. (Toronto: Carswell, 1992): 106

INTRODUCTION

1Pursuant to subsection 38(1) of the Human Rights Code [R.S.O. 1990, c. H.19] (the "Code"), we were appointed on June 10, 1993, by the Minister of Citizenship as a panel of persons appointed under subsection 35(1), to form a Board of Inquiry (the "Board") to hear and decide the complaints of Dawn Kearney, J.L., and Catarina Luis (the "complainants") alleging discrimination in accommodation on varying grounds by Bramalea Ltd., The Shelter Corporation, and Creccal Investments Ltd. (the "respondents"), respectively.

2Ms. Luis' complaint alleged that the use of an income criterion amounts to discrimination on the basis of sex, marital status, citizenship, place of origin, family status, and receipt of public assistance. J.L.'s complaint alleged that her application for an apartment was denied because of the application of an income criterion. She alleges this constitutes discrimination on the basis of age and sex. She also alleges that income criteria have an adverse effect on single women (marital status) and young people (age). Ms. Kearney's complaint alleged that Bramalea used an income criteri[on] and the use of that criterion constitutes discrimination on the basis of age and sex.

3While the grounds of discrimination vary depending on the personal characteristics and situation of the complainants, the nature of the alleged discriminatory act is the same in all three instances. At issue is the respondents' use of income criteria to deny accommodation to the complainants.

4The Commission's position is that the application of the criteria constitutes adverse effect discrimination, because it is the application of a rule that results in the exclusion or the restriction or preference of a number of designated groups. The complainants' position is that this case involves systemic discrimination and the central issue is whether landlords are permitted to exclude protected groups using a mathematical formula, that is, a rent-to-income ratio. Further, counsel for the complainants, Ms. Symes, alleged that alternate measures to that of income criterion such as the use of credit checks, co-signers, and previous rental history also violates the Code. Counsel did not argue that the landlords applied the income criteria differentially, based on personal characteristics, but alleged instead that the use of income criteria resulted in direct discrimination for some protected groups.

5Counsels for the respondents Shelter Corporation and Bramalea, Mr. Goudge and Mr. Doane, agreed that an income criterion was applied in these cases. However, the respondents maintained that the issue is "whether or not the Code prohibits credit grantors, such as landlords, from determining who they should extend credit to on the basis of their ability to pay". It is their position that renting an apartment is in the nature of a credit transaction. Furthermore, they argued that the use of minimum income criteria does not discriminate or select on the basis of any of the prohibited grounds under the Code. The respondents take the position that income criteria is bona fide and that without the ability to screen tenants in this manner and thereby limit their risk, landlords would suffer undue business hardship.

6Counsel for the respondent, Creccal Investments, disputed that Ms. Luis had any ground s for a complaint. Counsel said that on the date in question, there was no apartment available to rent to Ms. Luis; therefore, Creccal could not have discriminated against her. Creccal did not call evidence on the issue of whether the use of an income criterion was reasonable and bona fide, and relies on the evidence and submissions of counsel for Bramalea and Shelter on this issue.

PRELIMINARY MOTIONS

Interveners

7A number of organizations (the "applicants") sought standing before the Board as interveners. Counsel for the complainants argued that the Board should consider what type of intervener status the applicants should be given. The issue was whether interveners should be treated as parties with all the inherent rights and obligations, or as "friends of the Tribunal" with the participation rights being determined by this Board.

8Counsel for the Commission, the complainants and the respondents agreed in their submissions that the applicants should not be added as parties as they do not meet the requirements in s. 39 of the Code. While the applicants in their submissions did not ask to be added as a party to the complaint, they indicated that they would like to be considered as friends of the Tribunal and as such requested a wide range of participation, from adducing evidence with rights similar to a party, to making written submissions at the close of the case. Counsel for the parties agreed in their submissions that the applicants should be granted intervener status as friends of the Tribunal, but urged us to restrict the interveners to written submissions on the evidence.

9After considering the submissions of counsel for the parties and the applicants, we granted intervener status to the applicants. However, we restricted their participation in the process: all interveners were permitted to submit a written brief, based on the evidence adduced at the hearing. We indicated that any brief not based on evidence before the panel would be of little assistance to us.

Motion for an Adjournment of the Proceedings

10Counsel for respondents Bramalea Inc. and Shelter Corporation, Mr. Scott, moved for an adjournment of the proceedings on April 13, 1994, arguing that the Commission and complainants had not provided complete disclosure of their case. He asked for an adjournment of dates set in June and July 1994. Mr. Scott set forth the grounds for the motion as follows:

1. The Ontario Human Rights Commission has failed to provide complete disclosure of its case. The report of Dr. Hulchanski was only recently received, and is not complete. We have also been advised by Mr. Griffin that the Commission will be relying on further expert reports that it has yet to produce. Even if the Commission were to make complete disclosure immediately, there would be insufficient time for us to prepare the respondents' defence.

2. The complainants have provided virtually no disclosure of their case. Even if they were to make complete disclosure immediately, there would be insufficient time to prepare the respondents' defence.

11In support of his motion to adjourn, Mr. Scott cited R. v. Stinchcombe, 1991 CanLII 45 (SCC), [1991] 3 S.C.R. 326, and Christian v. Northwestern General Hospital [sub nom. Ontario (Human Rights Comm.) v. Ontario (Human Rights Board of Inquiry)] (1993), 1994 CanLII 10981 (ON CTGDDC), 115 D.L.R. (4th) 279 [1993 CanLII 16421 (ON CTGDDC), 20 C.H.R.R. D/498] (Ont. Ct. (Gen. Div.)), a decision of the Divisional Court approving a board of inquiry decision which relied upon the reasoning of Stinchcombe. In Stinchcombe, Mr. Justice Sopinka held that:

The Crown has a legal duty to disclose all relevant information to the defence. The fruits of the investigation which are in its possession are not the property of the Crown for use in securing a conviction but the property of the public to be used to ensure that justice is done.

A new trial was ordered and all relevant information was ordered to be disclosed. In Northwestern General Hospital, the Board followed the reasoning in Stinchcombe, and ordered production of the Commission's witness statements and other documents related to the investigation. The Commission had refused to disclose names and statements of witnesses interviewed during the investigation to protect the confidentiality of the witnesses and to foster an environment where witnesses felt comfortable about coming forward with relevant and sometimes sensitive information. Furthermore, the Commission was concerned about the potential for intimidation of witnesses, who may have had continuing involvement with the respondents. The Commission also wished to guard against reprisals. In upholding the Northwestern General Hospital decision, the Divisional Court affirmed that the important principle enunciated in Stinchcombe is that [at p. D/500, para. 16]:

[J]ustice was better served when the element of surprise was eliminated from the trial and the parties were prepared to address issues on the basis of complete information of the case to be met.

It does not take a quantum leap to come to the conclusion that in the appropriate case, justice will be better served in proceedings under the Human Rights Code when there is complete information available to the respondents.

Mr. Scott argued further that it would be a breach of natural justice to require the respondents to proceed with the hearings as scheduled.

12In opposing the motion of adjournment before us, counsel for the Commission replied that the Commission had already disclosed documents and witness statements to the respondents in compliance with the requirements for disclosure set out in Northwestern General Hospital. Mr. Griffin argued that in asking for further disclosure Mr. Scott was seeking a form of discovery. He argued further that under the Statutory Powers Procedure Act, R.S.O. 1990, c. S.22, as amended (hereinafter "SPPA") a respondent has a right under s. 8 to obtain particulars about the allegations. But the Commission does not have to provide the details of how it will prove its case. Mr. Griffin cited Dubajic v. Walbar Machine Products of Canada Limited (1980), 1980 CanLII 3911 (ON HRT), 1 C.H.R.R. D/228 (Ont. Bd.Inq.) in support of his argument. In Walbar, the respondent argued that pursuant to s. 8 of the SPPA, it was entitled to get more information from the Commission before the hearing started.

Counsel for the respondent in Walbar argued that [at p. D/228, para. 1997] s. 8 "required the Commission to go even further and disclose the evidence upon which it intended to rely in proof of its factual allegations". The Board in Walbar rejected this argument and held s. 8 does not contemplate a means of obtaining discovery of documents for inspection or statements of evidence by which it is intended to make out a case. It was Mr. Griffin's submission in the case before the Board that the respondent had known from the beginning the particulars of the case and had had full disclosure as required by Northwestern General Hospital.

13With respect to the respondents' claim of insufficient time to prepare, Mr. Griffin pointed out that the expert reports referred to by the respondents were provided two and a half months before the hearing. The other reports would be available a month before the hearing. Mr. Griffin estimated that the respondents would have about five months of preparation before their first witness testified.

14In opposing the motion for adjournment, counsel for the complainants, Ms. Symes, adopted the submissions of Mr. Griffin and indicated that she had provided Mr. Scott with a list of witnesses. It was not complete as it depended upon the Board's ruling with respect to interveners. Ms. Symes undertook to provide a complete report on the information to be given by expert witnesses at least a month in advance of the date when the experts would testify. She argued that this was consistent with the Rules of Civil Procedure, by which the Board is not even bound.

15In relation to production, Ms. Symes argued that the Code provides the board with very limited powers. Once appointed, s. 39 requires the board to hold a hearing. It outlines who the parties are, and how parties are added. Subsection 39(4) says:

Where the board exercises its power under clause 12(1)(b) of the Statutory Powers Procedure Act to issue a summons requiring the production in evidence of documents or things, it may, upon the production of the documents or things before it, adjourn the proceedings to permit the parties to examine the documents or things.

16There is nothing else in the Code which gives the board powers to control proceedings. The board must then turn to the SPPA to determine its powers. Section 12 says:

A tribunal may require any person, including a party, by summons,

(a) to give evidence on oath or affirmation at a hearing; and

(b) to produce in evidence at a hearing documents and things specified by the tribunal, relevant to the subject-matter of the proceeding and admissible at a hearing.

17Ms. Symes argued further that in Northwestern General Hospital the Divisional Court divided the human rights process into two stages. The first stage is the investigation stage and the second occurs after referral to the board. In Ms. Symes' submission, the Divisional Court approved the production of information contained during the investigation stage and this has been provided to the respondents. Documents produced for the purpose of litigation after the board was appointed should be excluded since a litigation privilege attaches to those documents.

18After considering the submissions of the parties, the request for adjournment by the respondents was denied. The Board was persuaded by the submissions of the Commission's and complainants' counsels, the provisions of the Code and the SPAA do not give the board of inquiry a general power to order discovery as opposed to disclosure. The Code and subsection 12(1)(b) of the SPPA give the board the power to issue a summons requiring the production of documents and to grant an adjournment for their review. Section 8 of the SPPA also entitles the respondents to disclosure of the particulars about the allegations.

19The respondents were asking for more than disclosure. They were asking [that] expert reports from the Commission and complainants be served on them at least three months before the hearing even though several had already been delivered and others were promised one month in advance of the hearing.

20There is no provision in the Code requiring production of an expert's report three months prior to the hearing, although it is especially helpful in a complex case to provide expert reports well in advance of the hearing to avoid requests for adjournment. We were not convinced that there would be a breach of natural justice in going forward with the hearings scheduled through June and July if the respondents did not get the reports three months before the hearing. As Commission counsel indicated, given the length of the case it would be many days of hearings before the Commission's experts testified.

21We were satisfied that the Commission had provided disclosure consistent with the requirements in Northwestern General Hospital, and that the principles emphasized in Stinchcombe had been preserved.

22The respondents' motion for an adjournment was also based on the request for further disclosure of the complainants' case before proceeding. However, the request for disclosure was not specific. In her submissions, Ms. Symes indicated that she had provided a provisional list of witnesses. She argued, as did Commission counsel, that she was not required to hand over the evidence which she would call to support her case. The Commission had provided particulars and full disclosure of its case. We were not persuaded that the respondents had inadequate disclosure of the complainants' case. Nor were we persuaded that there would be a breach of natural justice in proceeding with the hearing.

23Although the Board denied the initial motion for adjournment, the hearing days scheduled for June and July were subsequently adjourned because of a serious illness of a counsel, and the hearing did not commence on the merits until December 5, 1994.

MOTIONS TO AMEND THE COMPLAINTS

24At the beginning of the hearing on the merits of the complaints, we made several rulings at the request and on the consent of the parties. Thus, we granted the motion by counsel for the Commission to amend the complaint of Catarina Luis dated May 4, 1992, against Creccal Investments to include discrimination on the ground of race. Counsel for the respondent, Creccal Investments, did not object to the amendment on the basis that the amendment is related to an allegation of systemic discrimination on the ground of race, and that there is no allegation of direct discrimination because of race.

25We granted the motion by counsel for the Commission to remove the personal respondents from all the complaints. Finally, we granted the motion by counsel for the Commission that the complainant J.L. be referred to as J.L. in all public statements including the Board's decision. The Board directed anybody who intends to report on this case, either to their own clients or to anyone else, to refer to this complainant as "J.L.".

THE EVIDENCE

26The Board held over sixty days of hearings, over a period of three years. Given the extensive evidence tendered, we have summarized the evidence upon which we rely and, where necessary, to explain the reasons for our decision. Some of the evidence will only be referred to in our reasons for our decision. Much of the evidence of the individual complainants is not disputed.

Dawn Kearney

27Ms. Kearney was 17 years old at the time of the complaint and had just completed Grade 10. She was married and living with her husband's parents and his sister in a three-bedroom apartment in a building owned by Bramalea. She was pregnant. Initially, her husband was earning $7.24 an hour. When this was increased to $9.24 an hour, Ms. Kearney and her husband decided to look for an apartment of their own. Their area of search was Brampton as they were familiar with the surroundings, and family and friends lived in the area. They decided they would be able to afford rent between $550 and $650 per month. They arrived at this amount because Ms. Kearney's in-laws were paying just over $650 plus parking for a three-bedroom apartment.

28In the middle of September 1988, Ms. Kearney and her husband, Michael, went to Bramalea's office in Brampton. They met with a woman named Christine and asked for a two-bedroom apartment in the range of $600. Ms. Kearney said Christine asked if she was working, what her husband's job was, how long he had been employed there, and how much he made. She testified that she was told by the rental agent that they would have to make $30,000 a year in order to qualify for an apartment. The rental agent also mentioned that they were too young. Ms. Kearney's application for an apartment was denied.

29After having been rejected for an apartment, Ms. Kearney and her husband continued to live with her in-laws. Within weeks, she gave birth. Ms. Kearney described living with her in-laws as being very crowded. There was a lot of tension and arguments. There was no privacy. In the end, Ms. Kearney's in-laws left the apartment and Ms. Kearney, her husband, and her daughter occupied the apartment effective April 1, 1990. The amount of rent was $655 per month which was increased to $687 to include parking. Rental cheques made out to Bramalea from May 1, 1990, to November 1990, were submitted as evidence that Ms. Kearney and her husband paid their rent. The Kearneys vacated the apartment at the end of February 1991.

Catarina Luis

30Ms. Luis is a refugee from Angola. She left Angola after her husband, who was affiliated with the opposition party in Angola, disappeared. When she arrived with her two-year-old daughter in Canada on January 1, 1988, she spoke no English. First she stayed at a battered women's centre and then at a Salvation Army family resource centre in Brampton. After a series of temporary accommodations, Ms. Luis found a basement apartment in Toronto and moved in on November 1, 1988. The rent of $550 a month was reduced by the landlord to $500 after she told him of her difficult circumstances, being a refugee and single mother. The rent was increased a year later to $523. Ms. Luis was not satisfied with the conditions of the apartment: it was damp in the winter and very hot in the summer. There was also a problem with insects and mice. As a result, in the summer of 1990 Ms. Luis started to look for apartments through newspaper advertisements, friends, and by walking around and looking for "for rent" signs. She found an apartment that she was very interested in at The Crossways. Ms. Luis produced a newspaper clipping, which was verified as coming from the August 17, 1990, edition of the Toronto Sun. The clipping read: "At Dundas West Subway Bachelor at $585, 1 bdrm. at $760, util. incl. Pool, gym, air, laundry, shops, lakeview. 2340 Dundas West." This is the building known as The Crossways, owned by Creccal Investments.

31Ms. Luis testified that $700 a month was the maximum she could afford to pay for an apartment. She said that in order to arrive at this figure, she added her income (family benefits and earnings from a part-time job) and subtracted her cash costs.

32Ms. Luis testified that she called The Crossways and asked if the bachelor apartment advertised was still for rent and was told that it was still available. She went to the Crossways with her then four-year-old daughter arriving approximately an hour after the phone call. The receptionist told her to wait for the rental agent. While Ms. Luis was waiting, she filled out a "guest card" which requires disclosure of individual yearly income.

33Ms. Luis testified that she waited for about fifteen to twenty minutes and the rental agent, later identified as Ms. Gravelle, came in to see her. Ms. Luis asked to see the bachelor apartment that was for rent, but Ms. Gravelle said that the apartment was not available. Ms. Luis testified that she told the rental agent that she was told an hour before that it was available. Ms. Luis testified that Ms. Gravelle asked how much she made to which she replied between $1,500–$1,600 a month. Ms. Gravelle then told her that it was not enough. She was told that she had to make at least $2,000 a month in order to rent the bachelor apartment. This was a policy of the landlord. Ms. Luis testified that she called The Crossways soon afterwards without identifying herself to see if the bachelor apartment was still available and was told it was.

34The only evidence in dispute is whether there was an apartment available when Ms. Luis applied to Creccal on August 17, 1990. Evidence of the respondents was that there was no apartment available on that day. However, the evidence of Ms. Luis was that she was told there was an apartment available and the newspaper advertisement stated this and that is why she attended at The Crossways. It does not matter whether there was a vacant bachelor available on August 17, as the evidence indicated that there was constant turnover of apartments at The Crossways. Mr. Luigi Di Geso, an employee of Creccal, gave evidence that from August 17, 1990, to August 31, 1990, seven bachelor apartments and one junior bedroom apartment were available for rent between September to November. These apartments were all within the cost that Ms. Luis had determined she could afford to pay. We have no doubt that if Ms. Luis had met the respondents' income criteria she would have received one of the next vacant apartments.

35Ms. Luis said that after she was turned down by The Crossways, she continued to live in the basement apartment for a little more than a year and finally rented a bachelor apartment downtown. Ms. Luis provided the Board with copies of her rent payments in the amount of $630 a month, paid over the next year. She had to come up with first and last months' rent in the amount of $1,260 when she first moved in. Ms. Luis described the conditions in the building as not being desirable nor having the amenities that she saw advertised at The Crossways. The apartment she wanted to rent at The Crossways was $585 per month.

36Ms. Luis currently rents an apartment for $690 per month. She is on social assistance and her benefits total $1,469 per month. Her rent is 47 percent of her income. Ms. Luis testified that her first priority at the beginning of the month is to pay the rent. She then described various coping methods she uses to deal with her income and expenses: walking or taking public transit instead of taxis; rarely buying clothes for herself or her children and buying all clothes as cheaply as possible; going to food banks or to the Goodwill store.

J.L.

37J.L. was not called as a witness. Ms. Marta Dickinson, referred to in J.L.'s complaint as Marta Ready, was called. Ms. Dickinson said that she and J.L. were interested in renting an apartment together. They were looking for a two-bedroom apartment in the Meadowvale area. J.L. wanted to be close to her school and Ms. Dickinson was working at Meadowvale Town Centre. This was also close to Ms. Dickinson's family and friends. They had not worked out how much rent they felt they could afford before setting out on their search. She was 18 at the time and J.L. was 16. Ms. Dickinson was working at Kinney Shoes and being paid $6.50 per hour plus commission. Her pay cheque was about $625–$675 net every two weeks.

38Ms. Dickinson said that after finding an advertisement in the newspaper for a place called Aquitaine Shores, which is owned by Shelter Corporation, she and J.L. went to the rental office of Aquitaine Shores on June 1, 1990, and met with a rental agent. The rental agent showed them two available apartments. Both apartments cost $906 per month. They chose the apartment with beige carpet and were told by the rental agent that it would be available June 9, 1990. They returned to the rental office and filled out application forms. Ms. Dickinson said that the rental agent asked how old they were and how much money they earned. She said the rental agent seemed surprised by their ages.

39Ms. Dickinson said that she was told by the rental agent that she needed a letter from her employer, which she obtained, and that they needed to have first and last month's rent. They understood from the rental agent that if they met those conditions, they had the apartment as of June 9, 1990. It was a Friday when they filled out the application, and Ms. Dickinson said she was told that they had until Monday to bring in the employer's letter and the money. On Monday, they complied with both requests.

40Ms. Dickinson went in on Monday and gave the rental agent a cheque for $906 and a letter from her employer. She said that she and J.L. had agreed that one of them would cover the first month's cheque and the other, the last month's cheque. Ms. Dickinson said that J.L. delivered her cheque for $906 the same day.

41Ms. Dickinson said that two days later, she went to Aquitaine Shores to show one of her friends where she would be living. The rental agent told her to see the property manager and she did. Ms. Dickinson was told by the property manager that she and J.L. did not qualify for the apartment. The property manager gave her back her application and the cheque, and told her that they weren't making enough money. Each one of them had to be making enough money to cover the month's rent. The property manager said specifically that they had to be making over $32,000 a year each. She said that the application form along with their two cheques were returned to her.

42Ms. Dickinson was shown their application for tenancy which contained their birthdates, her employer, and her annual income of $16,000 from two part-time jobs. Under cross-examination, Ms. Dickinson admitted that this amount was a rough estimate of her earnings. There is a line through the application stating "declined–insufficient income and no credit references, June 6, 1990".

THE EXPERTS

John Stapleton

43Mr. Stapleton was qualified as an expert in social assistance policy development and was called to give evidence with respect to the evolution and development of social assistance programs in the province of Ontario. Mr. Stapleton provided the Board with his "Report on Social Assistance Programs in Ontario". His report included the historical development of social assistance which we will not summarize here but which was helpful in understanding the historical framework of Ontario's care of its poor and how that shapes the attitudes that society has towards the poor today. We were persuaded by Mr. Stapleton's evidence that these attitudes result in discrimination against the poor who are seeking accommodation.

44Mr. Stapleton also described how social assistance is obtained in Ontario under the [Family Benefits Act R.S.O. 1990, c. F.2] ("FBA"), and the General Welfare [Assistance] Act [R.S.D. 1990, c. G.6] ("GWA"). Various rules and financial testing exist under both regimes, but Mr. Stapleton commented that GWA financial testing is more stringent than under FBA. The GWA, commonly known as welfare, is generally a short-term program. Persons collecting GWA are generally those who are unemployed employable persons, those who have exhausted their employment insurance benefits or were not eligible for employment insurance.

45Mr. Stapleton said the most recent changes to social assistance benefits came about as a result of a report of the Social Assistance Review Committee called "Transitions". This 1988 report made 274 recommendations in social assistance and related areas. According to the report:

... in most larger urban centres, rental accommodation can consume 40 percent to 70 percent of monthly benefits. There is general acceptance that households should devote no more than 30 percent of income to cover shelter costs if they are to retain sufficient resources to purchase other basic necessities required to operate and maintain a household.

The report further provides:

While a 30 percent affordability threshold may be acceptable for single people, many housing analysts would support the use of even a 20 percent or 25 percent affordability threshold for families. The reality however, is that in Toronto . . . a substantially higher proportion of recipients (were) paying more than 50 percent of their incomes on rent.

46Mr. Stapleton also explained how refugee claimants became eligible for social assistance.

In a Divisional Court decision in 1984, the province was found to be erring in law in terms of when it did not allow refugee claimants to be eligible for assistance because they were not landed, and therefore not residents of Canada. The social assistance . . . administrators then were told that they had made an error by excluding them on residency grounds.

Mr. Stapleton explained that the reasons why so many refugee claimants are on social assistance is because they are unsponsored. Sponsored immigrants have Canadian sponsors who contract with the federal government to provide them with the necessities of life. Further, in the mid-1980s the federal government decided for policy reasons that work permits would not be provided to refugee claimants, so a very high proportion of refugee claimants had to go on social assistance. This policy was changed in early January 1994 and refugees are now allowed to apply for work permits.

47Mr. Stapleton described being poor in Ontario as not having enough money to be able to meet the basic necessities of life: food, clothing, shelter and personal care.

48Mr. Stapleton addressed some of the myths that have arisen concerning persons on social assistance. One myth is that people on welfare move a lot. Mr. Stapleton testified that there is no evidence that people on assistance move more frequently than anyone else. Another myth is that social assistance recipients are irresponsible with their money. Mr. Stapleton said that given the average amount that a recipient has to spend on shelter and the high number of recipients who have to use food banks, it is almost impossible to spend the money irresponsibly as it is applied to necessities. In addition, Mr. Stapleton said that some landlords believe that persons in receipt of public assistance will not take care of the (landlord's) property, that the property will be abused. Some believe that they will not be paid on time. Some believe that social assistance is not a secure form of income, and that the person may not have the money to pay for the accommodation or use the money for other things and neglect to pay their rent. Mr. Stapleton said that the impact of these myths is that social assistance recipients have a harder time finding accommodation.

49Mr. Stapleton testified that vacancy rates also have an effect on a recipient's ability to obtain accommodation. When the vacancy rate is low, recipients have a harder time obtaining housing. When the vacancy rate is high, there is a greater possibility of securing housing.

Chandra Pala

50Mr. Pala is an expert in both statistics and social assistance policy analysis. Mr. Pala provided the Board with his "Report on Ontario's Social Assistance Rate Structure". He gave evidence on the policy of the government, the statistics with respect to the profile of social assistance recipients in Ontario, and a profile of low income people in Ontario and the cost of housing for both groups. The following is a summary of the evidence which was helpful to the Board in assessing the impact of income criteria on people in receipt of social assistance.

51Mr. Pala reviewed the various income support programs available at the federal and provincial levels of government, and how rates for social assistance are set in Ontario. He then explained that there are two major components of social assistance benefits, the basic allowance and the shelter allowance.

52Mr. Pala testified that in 1989, the Social Assistance Review Committee found that shelter allowances were inadequate and complicated. A variety of recommendations were made. Using the example of a sole-support parent with one child under 9, Mr. Pala explained how the benefits would be calculated: basic allowance for the adult is $268 and for the child it is $218 for a total of $486 per month. Furthermore, there is now a "basic shelter" allowance which for a family of two is $185 which would make the total $671. With respect to the shelter costs, any costs in excess of $185, would be paid up to $550 of shelter cost or $365 of variable shelter allowance. In other words, the minimum is $185 and the maximum is $550. The maximum monthly allowance this family would get is $1,036 per month, an increase from the rate paid before the reform.

53Mr. Pala testified that given the rate structure for assistance, if a landlord required an annual income of $22,000 for an apartment which costs $585 to rent, no one on social assistance would meet that qualification.

54Mr. Pala provided the Board with statistics available through Ministry of Community and Social Services, on the breakdown of persons who are on social assistance. In 1990, 37.3 percent of recipients were sole-support parents. In 1994, 33.4 percent were sole-support parents. Based on 1990 statistics, Mr. Pala outlined in his report the type of accommodation persons receiving assistance were living in:

· 6 percent of recipients are home-owners,

· 14.2 percent are in public housing such as that which is provided by the Ontario Housing Corporation and municipally owned housing,

· 61.1 percent of recipients rely on housing that is available through private for-profit landlords.

The reliance on private, for-profit housing increased to 71.8 percent in 1994.

Michael Ornstein

55Dr. Ornstein is a sociologist and statistician, and was qualified as an expert witness to provide the Board with his evidence and report "Income and Rent: Equality Seeking Groups and Access to Rental Accommodation Restricted by Income Criteria". Dr. Ornstein used the term "equality seeking groups" to identify people protected by the Code. Dr. Ornstein addressed issues related to the differences in income of protected groups and the general population and access of protected groups to housing. He also addressed the question of whether having less income means more percentage of income must be spent on rent.

56The data contained in Dr. Ornstein's report come from a variety of sources including census data from the Toronto Metropolitan Area ("CMA"). Dr. Ornstein also provided figures based on the census for Brampton (where Ms. Kearney was attempting to rent an apartment from the respondent Bramalea) and compared them to the Toronto CMA figures. He also provided the Board with analysis on Ontario-wide patterns.

57Dr. Ornstein's analysis of the data contained in his report and in his testimony was of assistance to the Board in understanding the impact on protected groups when landlords use income criteria. A summary of his evidence is as follows.

(a) Income Level of Equality Seeking Groups

58Lone Parents

· 90 percent of lone or single parents are female

· of the female lone parents who rented, 51 percent had incomes of less than $25,000

Age

· about two-thirds of all unattached persons are under the age of 20

· 68 percent of unattached women under the age of 20 have less than $10,000 in income

· 96 percent of unattached women under the age of 20 have less than $25,000 in income

· of couples in a two-person household under the age of 20, 16 percent of those who have female partners have incomes of less than $10,000. 20 percent of those who have male partners have incomes of less than $10,000

· of couples in a two-person household under the age of 20, 44 percent of those who have female partners have incomes of less than $25,000. 50 percent of those who have male partners have incomes of less than $25,000

Race

· of unattached women, 34 percent of first nations women, 22 percent of black women, 23 percent of South Asian women, 23 percent of East and Southeast Asian women, and 35 percent of West Asian women have less than $10,000 in income. 13 percent of non-minority women have incomes of less than $10,000

· of unattached women, the figures for those who have incomes of less than $25,000 is: 73 percent of first nations women, 70 percent of black women, 70 percent of South Asian women, 73 percent of East and Southeast Asian women, 82 percent of West Asian women, and 56 percent of non-minority women

Citizenship, immigration status and place of origin

· non-citizens have less income than citizens

· among female lone-parents, non citizens are roughly 5 percent more likely than citizens to be in the lowest income categories

· non-permanent residents (which include refugees as well as people with student and work permits) have the lowest incomes

· people born in Canada have the highest incomes

· for unattached women, 58 percent of the women from Africa have incomes below $20,000 compared to 41 percent of women born in Canada

Dependence on social assistance

· the Census provides information on the amount of "other government income" received in a year

· of unattached women who receive government income, approximately 92 percent have under $20,000.

(b) The Exclusionary Effect of Income Criteria (30 Percent) on Equality Seeking Groups

59The methodology used by Dr. Ornstein to determine the exclusionary effect of income criteria on equality seeking groups is contained in his report. Dr. Ornstein divided rents into three categories, "low rent", "very low rent", and "median rent" accommodation. His conclusions are as follows:

· one half of all unattached women are unable to meet the 30 percent income criterion applied to low rent accommodation

· of female lone parents, 51 percent are unable to meet the criterion for "low rent" accommodation

· 92 percent of all unattached women under the age of 20 do not have sufficient income to qualify for low rent accommodation

· 77 percent of couples in which the male partner is young (under 20 years of age) do not qualify for low rent accommodation

· members of racial minorities are less likely to have access to rental accommodation restricted by income criteria

· citizens, people born in Canada, and less recent immigrants have higher incomes and are more likely to meet income criterion than recent immigrants and non-permanent residents

· those who are more dependent on government income are less likely to meet income criterion.

60Dr. Ornstein concluded that :

Young unattached people, young lone parents, and unattached people who are dependent on social assistance are almost completely excluded from accommodation restricted by income criteria. All recent immigrants, not only refugees whose position would surely be still worse, are also denied access to most rental housing restricted by income criteria. Other equality seeking groups, including visible minorities, non-citizens, and people born outside Canada also have significantly less access to restricted accommodation ...

(c) The Actual Rent-to-Income Ratios of Equality Seeking Groups

61Dr. Ornstein also provided this Board with pertinent information on the actual situation of renters, that is, the ratio of income equality seeking groups pay now on rent. Dr. Ornstein's data is as follows:

· between one quarter and a third of all tenants pay 30 percent or more of their income on rent

· of those who pay more than 30 percent of their income on rent, they include: 37 percent of all unattached women; 47 percent of all female lone parents; and 22 percent of all couples in two-person households

· the lowest income groups devote dramatically more of their resources to rent

· female lone parents with under $10,000 in household income pay 89 percent of that income on rent; those under $20,000 in household income pay 64 percent of their income on rent; and for those under $30,000 in household income pay 43 percent of their income on rent

· unattached women with under $10,000 in personal income pay 70 percent of their income on rent; those under $20,000 in personal income pay 83 percent of their income on rent; and for those under $30,000 in personal income pay 50 percent of their income on rent

· if every landlord in the CMA imposed a 30 percent income requirement in order to rent units in their apartment building, between a third and a quarter of them would be homeless and landlords would be dealing with 25 percent–33 percent of the units being vacant

62Dr. Ornstein concluded the fact that "poor people devote so much of their income to rent is neither voluntary, nor the result of poor apartment-hunting; rather it reflects the price of housing, which is a necessity".

(d) Comparisons Between Toronto and Brampton

63It was Dr. Ornstein's opinion there was little difference between the income and accommodation situation in the Toronto CMA and in Brampton. Brampton is different in a number of respects in terms of the population and cost of housing, but the basic patterns are extremely similar.

The similarity between Brampton and Toronto arises from the continuity in the housing and labour markets in the two locations, which is brought about by the high levels of mobility in the larger urban centres surrounding the Toronto CMA. These results also show that Dawn Kearney's experience of being denied accommodation because of income criteria is typical of the effect of such criteria on equality seeking groups in the community of Brampton.

(e) Ontario-Wide Patterns

64Dr. Ornstein reported on whether the patterns observed in Toronto and Brampton prevail across Ontario. For this analysis, he divided the province into six geographic categories. The three largest CMAs in Ontario, Toronto, Hamilton and Ottawa, were treated separately, and the remainder of the province was divided into three categories: rural areas and urban areas with less than 100,000 population; urban areas with 100,000 or more population, excluding CMAs; and the combination of the four smaller CMAs, Kitchener-Waterloo, London, St. Catharines, and Windsor.

65Dr. Ornstein determined that in all six areas, female lone parents are the most likely to be paying 30 percent or more of their income in rent. Further, in all the locations, couples are the least likely to pay 30 percent or more of their income in rent. The broad patterns of household income, rent-to-income rations, and differentials in the ability to qualify for housing restricted by income criteria, are very similar. Finally, that there is regional variation in Ontario, reflecting segmentation in the labour and housing markets and demographic differences between regions, but the general patterns, such as the disadvantaged position of female-headed lone parent families, young people, and recent immigrants, are the same. Dr. Ornstein said:

Nearly one-third of all tenants in Ontario and even larger proportions of the tenants in equality seeking groups actually pay more tha[n] 30 percent of their income for accommodation. Were all landlords to apply income criteria, about one-third of all tenants in Ontario would have to move to other housing.

(f) Exclusion of Equality-Seeking Groups from the Respondents' Apartments

66Dr. Ornstein examined the situations of the three complainants in terms of rent levels and income criteria.

67Ms. Luis was told that she had to make at least $2,000 a month when she attempted to rent an apartment at The Crossways (the building owned by the respondent Creccal Investments). This amounts to an annual income of $24,000. Dr. Ornstein used the figure of $23,400 in his report as annual income. When looking at the income levels of equality seeking groups, Dr. Ornstein concluded that "... on all the grounds cited in Ms. Luis' complaint, the rent levels and income criterion in force at The Crossways systematically disadvantage equality seeking groups".

68J.L. alleged discrimination on the basis of sex and age when she and a friend, Ms. Dickinson, attempted to rent an apartment at Aquitaine Shores (the building owned by the respondent Shelter Corporation). They were not allowed to combine their incomes in meeting the criterion. Dr. Ornstein concluded that 98 percent of unattached women who were under the age of 20 did not have sufficient income to qualify as tenants. "Thus young unattached women are virtually excluded from Aquitaine Shores."

69Ms. Kearney alleged discrimination on the basis of sex and age when she attempted to rent an apartment at a building owned by the respondent Bramalea Inc. Dr. Ornstein concludes that "... the income criterion employed in the Bramalea building differentially excludes young people".

(g) Present Tenants in the Three Respondents' Apartments

70Dr. Ornstein conducted a survey in June 1994 of the tenants who currently rent at Aquitaine Shores, Bramalea, and The Crossways to determine the actual representation of tenants in these buildings. Dr. Ornstein used the income criterion cited in the complaints, 25 percent, 30 percent and 36 percent, and applied them to the corresponding buildings. There was a 50 percent response rate across all three buildings. Dr. Ornstein found that at all three locations there were substantial numbers of tenant households whose income is too low to meet the various income criteria: 7 percent of The Crossways tenants, 30 percent of the Bramalea tenants, and 13 percent of the Aquitaine Shores tenants have incomes below $20,000. He also determined that the representation of low income households in the apartments could not possibly have resulted from tenants moving into the building and having their economic circumstances change afterwards. Further, Dr. Ornstein reported that at The Crossways, 41 percent of the apartments are rented by households in which rents exceed 30 percent of the household income. At Aquitaine Shores, 25 percent of the households exceed a 30 percent income criterion and 17 percent exceed 40 percent (the criterion in J.L.'s case was 36 percent). At Bramalea, 50 percent of the households were beyond the 25 percent criterion. Dr. Ornstein concluded that:

Even allowing for a great deal of imprecision in these findings, the present distribution of incomes in the respondents' buildings is not compatible with the use of the income criteria that were employed at the time the complainants sought housing ... In light of these findings, one of our purposes in conducting the survey, to determine whether equality seeking groups were effectively excluded from the buildings as a result of the use of income criteria, no longer makes sense. It appears that the increased vacancy rate that followed the onset of the recession forced landlords to choose between using income criteria to assure desirable tenants and having vacant apartments. Given this choice, landlords appear to have abandoned the rigid application of income criteria as a means of selecting tenants.

71Dr. Ornstein's evidence persuaded the Board that there is ample evidence to conclude "that income criteria differentially affect equality seeking groups, defined on the basis of sex, marital and family status, age, citizenship, race, immigration status, place of origin and being in receipt of social assistance". Dr. Ornstein's evidence on cross-examination was consistent with his evidence-in-chief.

David Hulchanski

72Dr. Hulchanski was qualified as an expert in Canadian housing policy, rental housing policy including housing programs and problems, and housing affordability to give us his expert opinion and report entitled "Discrimination in Ontario's Rental Housing Market: The Role of Minimum Income Criteria". Background paper No. 1 — "How Households Obtain Resources to Meet their Needs: The Shifting Mix of Cash and Non-Cash Sources" defines income and how it relates to the calculation of income criteria. Background paper No. 2 — "The Use of Housing Expenditure-to-Income Ratios: Origins, Evolution and Implications" addresses, as the title indicates, the origins, evolution, and implications of using the housing expenditure-to-income ratio.

73Dr. Hulchanski drew a link between the concept of housing affordability and the use of income criteria. Minimum income criteria are used to determine the threshold of affordability. In other words, the inference is that if one pays over what one can afford, default in the rent is more likely.

74Dr. Hulchanski called the minimum income criteria "a rule of thumb". His research did not reveal when the practice of using this rule in the residential rental market began nor when its use became so widespread. "In Canada the 20 percent rule lasted until the 1950s when somehow the 25 percent rule came into use, only to be replaced in the 1980s by the 30 percent rule."

75Dr. Hulchanski's background paper No. 2 traced the origin and evolution of the minimum income criterion which has its roots in the nineteenth century and became part of the conventional wisdom associated with housing during the first half of the twentieth. It was used mainly by both the private and public sectors in an attempt to describe average household expenditures on housing. According to Dr. Hulchanski, there was little analysis of the ratio as a valid and reliable measure. The ratio became an established "rule" providing a summary description of the maximum proportion of income that should be devoted to rent or to a mortgage.

This assertion that rents or mortgage payments should not exceed a certain percentage of income permeates housing discussion throughout the 20th century. For some, the housing expenditure-to-income ratio became a measure of housing affordability for public policy purposes — a rule used in allocating public housing units. For others, it became a measure of ability to pay — a rule used in making business decisions (decisions about who to rent to and who to grant a mortgage to). Yet, as this history of the origins and evolution of the measure has found, it is difficult to find the users of this statistical average "rule of thumb" offering any empirical rationale in support for its validity and reliability as a measure. There is no study or set of studies or body of literature concluding that a certain ratio or any ratio is an appropriate measure of what a household ought to spend or is able to spend.

76Dr. Hulchanski said there are six uses of the expenditure-to-income ratio. They are: description (describe a typical household's housing expenditure); analysis (analyze trends, compare different household types); administration (administer rules defining who can access housing subsidies); definition (define housing need for policy purposes); prediction (ability of a tenant household to pay market rent); and selection (select tenants who can afford to pay the rent). Dr. Hulchanski divided the list into two categories. In Dr. Hulchanski's opinion:

The first three uses — description, analysis and administration — can be considered quite valid and helpful when used properly by housing researchers and administrators. "Used properly" means that the research methods and the statistical analysis techniques are properly carried out — i.e., no significant methodological errors are made. This leads to valid and reliable descriptive and analytic statements about the housing expenditures of the different types of households being studied. This type of description and analysis of household expenditure patterns can also be helpful in defining administrative rules about eligibility for means-tested, as opposed to universal, housing programs.

The improper and inappropriate use of housing expenditure-to-income ratios leading to invalid and unreliable results, is due to a variety of theoretical and conceptual errors. Uses four, five and six — definition, prediction, and selection — are all invalid uses for they fail to measure what they claim to be measuring, even if the research methods and the statistical analysis techniques are properly carried out. The ratio is faulty when used to define housing need and to predict the ability of households to pay for housing due to a faulty conceptualization of the income part of the ratio . . . An additional conceptual problem arises because it applies a statistical average of a group of households to an individual household, leading to the problem of statistical discrimination.

77In Dr. Hulchanski's report and in his background paper, he outlined his reasons for his opinion that the ratio is not a valid measure of housing affordability. The ratio only measures income, that is money income, and ignores other sources of support by which households meet their needs. Dr. Hulchanski explained that most households in Canada draw resources from the "five economies". There is the domestic economy, which is internal to a household, and is a form of self-provisioning (it is cheaper to cook for ourselves than go to restaurants which require cash). The second economy is called the informal economy, and includes extended family and close acquaintances, even co-workers, who provide all kinds of assistance (for example, a neighbour who is an electrician and makes a repair free of charge). The third economy all households draw on is the social economy. This includes the neighbourhood and community based agencies and groups, the voluntary sector, the charitable sector, non-profit sector, local social agencies, non-governmental agencies of various types. The fourth economy is the market economy, which is the major source of cash income. Finally, one can obtain resources from the government or state economy, such as social assistance as well as other kinds of assistance like job training, unemployment and old age pensions.

78In Dr. Hulchanski's view, these economies explain why people and/or households can spend a high percentage of their income on housing without being in severe trouble. In his opinion, the rent-to-income criterion mainly assesses the market economy. It considers the state economy to some extent. Further, reliance on other economies is the reason why the incidences of defaults on mortgages and rents are not high. It was his evidence that "... more than 30% of owners with mortgages and renters in both Ontario and Metropolitan Toronto spend at least 30 percent of their household income on housing costs. The Social Assistance Review Committee (1988) reported that in Toronto, Ottawa, Hamilton, and Waterloo, 70% or more of social assistance recipients were devoting 40% or more of their incomes to rental payments in 1986".

79In Dr. Hulchanski's opinion, if the housing expenditure-to-income ratio being used by the public and private sectors was a valid and reliable measure of affordability and ability to pay, these figures would indicate that the country was facing a huge default rate in rent and mortgage payments. In Dr. Hulchanski's opinion, this is an example of how income criteria does not stand up to empirical testing. It does not explain how it is possible that households (especially those with lower than average incomes and those in receipt of social assistance) can survive and continue to meet their housing needs when their shelter costs amount to so much more of their household income. In his view, the answer lies in recognizing the inadequacy of "household income" and "household budgets" as a measure of a household's ability to pay. Measures of ability to pay must recognize the multiple economies and shifting mix of resources.

80In cross-examination, Dr. Hulchanski consistently defended his opinion that the rent-to-income ratio serves no usefulness as a predictor of default. He reiterated that the failure of landlords to take into account the "five economies" results in the screening out of tenants who may be very good tenants. Dr. Hulchanski acknowledged that he could not prove nor disprove a link between income and default. Dr. Hulchanski emphasized that households draw on a variety of supports when a life event hits them.

81Dr. Hulchanski agreed on cross-examination that landlords do not appear to be uniform or absolute in the application of a 30 percent rent-to-income ratio. He also agreed that there is a point at which a household will have difficulty paying the rent. However, the household may not default: they could choose to move to a cheaper place.

82The Board was persuaded by Dr. Hulchanski's evidence that rent-to-income ratios, presently used by landlords, are not a valid criterion for assessing a person's ability or willingness to pay rent, nor that it predicts default. Nothing on cross-examination or in the respondents' evidence persuaded us that his opinion was wrong.

Maureen Callaghan

83The Board heard evidence from Maureen Callaghan, who is the housing coordinator with Jessie's Centre for Teenagers. Jessie's is a multi-service agency for pregnant and parenting teenagers. Ms. Callaghan submitted a report on the negative effects of minimum income criteria on pregnant and parenting teenagers. She gave evidence concerning the experiences and difficulties of pregnant and parenting teenagers in finding accommodation. Ms. Callaghan stressed the importance of housing in providing stability for a young person who is struggling both with becoming an adult and being a parent.

Sister Mary Jo Leddy

84Sister Mary Jo Leddy gave evidence concerning the experience and difficulties of refugees in obtaining housing from private landlords using income criteria. Sister Leddy is the executive director of Romero House, which is a resettlement house for refugees. The average stay of a refugee and his or her family is six to nine months. Romero House staff assist refugees in finding more permanent housing. Sister Leddy gave evidence concerning the difficulties of refugees in obtaining co-signers and credit references since they are new to the country. It is difficult for refugees to come up with first and last months' rent as they arrive in Canada with virtually no money.

Nancy Webb

85Nancy Webb is the executive director of Touchstone Youth Centre, which is a shelter for homeless youth. Ms. Webb talked about the difficulties that young persons have in finding shelter where an income criteri[on] is applied. Ms. Webb described the barriers experienced by homeless youth in obtaining housing: lack of rental history; lack of a guarantor; lack of money as they are often without a job. It was her evidence that as a result of their age, they are not seen by landlords to be stable or mature enough to rent an apartment.

Ann Fitzpatrick

86Ms. Fitzpatrick was qualified as an expert in child and family services and gave evidence with respect to the impact of the shortage of housing on children and their families. Ms. Fitzpatrick submitted a report entitled "Effects of Minimum Income Qualifications on Families with Children and Youth Seeking Housing". Ms. Fitzpatrick has worked over the past twelve years with the Catholic Children's Aid Society of Metro Toronto ("CCAS") and the Children's Aid Society of Metropolitan Toronto ("CASMT"). She has had extensive experience addressing the housing needs of families with children and youth leaving the care of the child welfare system.

87Ms. Fitzpatrick's evidence is that annually, the CASMT works with over 8,500 families and 18,500 children a year. Of the household population that CASMT provides service to, 56 percent of CASMT's clients are single parent, primarily female headed. Twenty-seven percent of the families served, and one in four children in care, are members of racial and ethnic minorities, many of whom are new immigrants or refugees. Fifty-six percent of families receive some form of social assistance. The vast majority of CASMT clients are tenants.

88Ms. Fitzpatrick quoted a statistic from Canada Mortgage and Housing Corporation that in 1990, the average income of single parents in core need was $14,200 and this group had an average rent-to-income ratio of 47 percent.

89CASMT attempts to prepare youth in care who are wards of the Society for future independence, including access to safe and adequate housing. $663 a month (prior to November 1994, the independence allowance was $567/month) is provided by CASMT to youth on independent living programs (living independently in the community). Youth most often access housing such as rooming houses, shared apartments with friends/family members, shelters, independent living group homes, and apartments in houses (e.g., basement apartments). CASMT allows youth to spend up to 70 percent of their allowance on rent.

90Ms. Fitzpatrick described the needs of families and the importance of optimizing the choice of housing available to meet those needs. The needs include the size of housing, adequacy of space, location, proximity to friends, family, support people, schools, and community. Ms. Fitzpatrick outlined some of the difficulties experienced by CASMT clients in renting apartments. Income criteria were a major factor in clients' difficulties in renting stable and safe housing although Ms. Fitzpatrick acknowledged that income criteria are not uniformly applied by landlords.

91Ms. Fitzpatrick also gave evidence concerning the impact on clients who have been refused rental accommodation of their choice. Families having to share accommodation with friends or relatives, parents sleeping in the same room or same bed as their children, increased noise and overcrowding — all these conditions lead to increased stress and health problems.

92Youth experience different problems than families in securing safe and stable housing. A co-signer or credit check is often required in addition to consideration of income criteria. Co-signers may be difficult to obtain in that the youth may have left home because of an abusive situation. Young persons may not have credit and are therefore unable to meet the credit check requirement. The inability to secure housing can have detrimental effects on youth which may range from homelessness to physical and mental health risks to the extremes of prostitution, malnutrition, or manic/aggressive anti-social behaviours.

Barry Lyon and Gary McIlravey

93Messrs. Lyon and McIlravey testified as a panel. They were both qualified as experts in residential market analysis, and gave evidence on the impact of rent arrears on the viability of residential landlords' business[es] at the hearing and in their report, "The Impact of Rent Arrears on the Viability of Residential Landlord's Businesses". The report assesses the degree to which tenant default [a]ffects the viability of operating a typical rental business. Mr. McIlravey testified that he was not aware of any studies or data that had established a link between default rates and rent-to-income ratios.

94The first part of the report described the nature of the rental business: the kinds of landlords, the demand environment for rental housing, the typical level of vacancy and how it was changed over time. The next part of the report dealt with how a rental business is typically evaluated, and determined to be viable or not. It compared the typical costs of doing business and the revenues achieved. The report reviews various sources to determine what the likely extent of tenant default is and how that affects the viability of the business.

95Messrs. Lyon and McIlravey concluded that the risk of tenant default is an insignificant factor in determining the viability of a residential rental business. They found that bad debt normally makes up less than 1 percent of gross revenue, including retail and wholesale businesses. They were of the opinion that there is no reason to suggest a greater risk of bad debt in the residential rental apartment business, than in others.

96They considered the effect of a typical level of bad debt on profitability and on return on investment and found that the effect was relatively insignificant. Further, they testified that eliminating an average level of bad debt altogether would only increase the rate of return by about one-tenth of one percent. Similarly, they found that doubling the average level of bad debt would only reduce the rate of return by one-tenth of one percent. A minor fluctuation in property tax rates, mortgage rates, or an unexpected repair bill, pose equal and potentially more serious risks for landlords than the risk of increased tenant default, in their view.

97Messrs. Lyon and McIlravey were of the opinion that rental businesses fail because the investor has paid more than the income can support or has failed to budget for necessary repairs, or has accepted a capitalization rate that is too low because of unrealistic expectations of equity appreciation. They concluded that tenant default on rent is not a significant cause of business failure in the residential rental business.

98Messrs. Lyon and McIlravey surveyed landlords who advertised apartments in the Renters' News and found that only 28 percent of landlords surveyed use minimum income qualifications. They found that the practice of using income criteria is most frequent among large property managers and least frequent among those landlords who are most vulnerable to tenant default — the small "mom and pop" landlords.

99It was their evidence that there is no reference to the use of income criteria in a widely used "Property Manager's Handbook" distributed by the Property Management Training Association. They were also of the opinion that there is no study showing any relation between rent-to-income ratios and default on rent. They agreed with the evidence of Dr. Hulchanski that the use of minimum income criteria to screen tenants seems to have evolved from commonly held prejudices without empirical research on which to base these practices.

100The data analyzed by Messrs. Lyon and McIlravey shows that tenant default is not a significant factor in determining the viability of a landlord's business. The average impact of default is 0.5 percent of gross income in larger buildings. They were of the opinion that with vacancy rates in the range of 1.8 percent, the cost of vacancies to landlords is a more significant factor.

101The Board was persuaded by Messrs. Lyon's and McIlravey's conclusions that:

Consequently, restricting applicants to apartment buildings on the basis of income in the hopes of reducing default, may be counter-productive to the landlord. In fact, this practice potentially creates additional costs to the landlord, by restricting demand and increasing vacancy, rather than creating any significant savings in the area of bad debts.

The Evidence of the Respondents' Experts

102The respondents also called a number of expert witnesses to give evidence, which we will refer to as needed in our reasons for our decision. After careful review of the respondents' evidence, we have decided that the respondents produced no evidence that the use of income criteria to screen tenants is a reliable predictor of default or the ability to pay rent. Furthermore, the respondents did not tender any evidence to support a finding that if landlords were not permitted to use income criteria to screen prospective tenants that they would suffer undue hardship.

DECISION

103In making its decision, the Board was mindful that the Supreme Court of Canada has consistently held that the proper interpretation of human rights legislation requires a broad, liberal, purposive approach. Most recently in Gibbs v. Battlefords and District Co-operative Ltd., 1996 CanLII 187 (SCC), [1996] 3 S.C.R. 566 [27 C.H.R.R. D/87 at D/94, para. 18], Sopinka J., citing earlier decisions, wrote as follows:

... This court has consistently held that human rights legislation is "fundamental" or "quasi-constitutional" and as such should be interpreted in a broad and purposive manner. As Lamer J. (as he then was) stated in Insurance Corporation of British Columbia v. Heerspink, 1982 CanLII 27 (SCC), [1982] 2 S.C.R. 145, at pp.157–58:

When the subject matter of a law is said to be the comprehensive statement of the "human rights" of the people living in that jurisdiction, then there is no doubt in my mind that the people of that jurisdiction have through their legislature clearly indicated that they consider that law, and the values it endeavours to buttress and protect, are, save their constitutional laws, more important than all others.

Sopinka J. also cited Zurich Insurance Co. v. Ontario (Human Rights Commission), 1992 CanLII 67 (SCC), [1992] 2 S.C.R. 321 [16 C.H.R.R. D/255 at D/263, para. 18]:

Human rights legislation is amongst the most pre-eminent category of legislation ... One of the reasons such legislation has been so described is that it is often the final refuge of the disadvantaged and the disenfranchised. As the last protection of the most vulnerable members of society, exceptions to such legislation should be narrowly construed . . .

104The meaning of the term "discrimination" was considered by the Supreme Court in Canadian National Railway Co. v. Canada (Canadian Human Rights Commission)(1987), 1987 CanLII 109 (SCC), 8 C.H.R.R. D/4210 at D/4227 (S.C.C.). The Court referred to the Abella Report, and the essential elements of systemic discrimination [para. 33248]:

Discrimination . . . means practices or attitudes that have, whether by design or impact, the effect of limiting an individual's or a group[s] right to the opportunities generally available because of attributed rather than actual characteristics . . .

It is not a question of whether this discrimination is motivated by an intentional desire to obstruct someone's potential, or whether it is the accidental by-product of innocently motivated practices of systems. If the barrier is affecting certain groups in a disproportionately negative way, it is a signal that the practices that lead to this adverse impact may be discriminatory. This is why it is important to look at the results of a system.

105Whether a case is one of direct or constructive discrimination, the burden of proof in all cases is the same. It is well established that it is the civil standard, a balance of probabilities (Ontario Human Rights Commission v. Borough of Etobicoke (1982), 1982 CanLII 15 (SCC), 3 C.H.R.R. D/781 (S.C.C.)). Further, in human rights cases, the burden of proof to establish a prima facie case rests with the Commission and the complainants (Ontario Human Rights Commission v. Borough of Etobicoke, supra; Ontario Human Rights Commission and O'Malley v. Simpsons-Sears Ltd., 1985 CanLII 18 (SCC), [1985] 2 S.C.R. 536 [7 C.H.R.R. D/3102]).

106A prima facie case is one which supports or proves the allegations made and which, if the allegations were believed, is sufficient to justify a decision in the complainant's favour in the absence of a response from the respondent. This applies to both direct and to adverse effect discrimination. The elements of a prima facie case of constructive discrimination are determined by the words of s. 11 of the Code. The components of a prima facie case as set out in Keene, Human Rights in Ontario, 2d ed. (Toronto: Carswell, 1992) at 126, were adopted in Thorne v. Emerson Electric Canada Ltd.(1993), 1993 CanLII 16439 (ON HRT), 18 C.H.R.R. D/510 at D/513–D/514 (Ont. Bd.Inq.):

(a) Existence of the factor — that the landlord uses a rent/income ratio or income criterion as a factor in assessing applications; that the use of such a factor is a "requirement, qualification or factor" as those words appear in s. 11 of the Code.

(b) Effect of the factor — that using the factor results in the "exclusion, restriction or preference of a group of persons who are identified by a prohibited ground of discrimination".

(c) Membership in the protected group — that the complainant (the applicant for tenancy with that particular landlord) is a member of the group (or groups if applicable) referred to in clause (b) above.

107In considering the second component above — the effect of the requirement, qualification or factor — it is clear that the Commission and the complainants need not prove that the factor excludes all people in the particular group. There is no requirement to prove that all members of a group share a particular characteristic. In Sehdev v. Bayview Glen Junior Schools Ltd.(1988), 1988 CanLII 8934 (ON HRT), 9 C.H.R.R. D/4881 (Ont. Bd.Inq.), constructive discrimination was established, even though the rule in question — a dress code that prohibited headwear — did not affect every Sikh or every Jew. A neutral rule in Janssen v. Ontario (Milk Marketing Board)(1990), 1990 CanLII 12505 (ON HRT), 13 C.H.R.R. D/397 (Ont. Bd.Inq.) was found to constitute adverse effect discrimination even though it did not equally affect every member of a particular faith.

108Cases of direct discrimination do not require a finding that all people bearing the relevant characteristic be discriminated against. The Supreme Court made this point again in Gibbs, as follows:

... The case law has consistently held that it is not fatal to a finding of discrimination based upon a prohibited ground that not all persons bearing the relevant characteristic have been discriminated against.

It would be illogical and contrary to the principles of statutory interpretation to have different standards for the "scope of the effect" of discrimination that depend on whether the discrimination was direct or constructive.

Catarina Luis v. Creccal Investments Ltd. ("Creccal")

(a) The Existence of the Factor

109As noted earlier in our findings of fact, Ms. Luis' evidence was that she went to The Crossways, owned by Creccal, to inquire about an apartment she had seen advertised. When she was asked how much she made, Ms. Luis replied that she made between $1,500 and $1,600 a month. Creccal's representative, Ms. Gravelle, said that it was not enough and that it was Creccal's policy that she had to make $2,000 a month to qualify for an apartment. When Ms. Luis pointed out what she was paying at her present apartment, Ms. Gravelle said that it did not matter.

110Ms. Gravelle testified that she had no recollection of meeting Ms. Luis. However, on the day that Ms. Luis looked for an apartment, the policy was that to qualify for a bachelor, the applicant needed an annual income of $ 22,000. Creccal does not run credit checks on people who do not meet the income criteria.

111Mr. Di Geso, the manager of The Crossways, testified that in 1990 Creccal used income criteria. He said that if an applicant was making $17,000 per year, The Crossways would explain that the applicant could not afford the apartment. Since Creccal required $22,000 per year for a bachelor apartment, an applicant making $17,000, $18,000 or $19,000 per year did not meet the income criteria. Creccal would not put the applicant on a waiting list if the income criteri[on] was not met.

112The evidence establishes that Creccal was using an income criterion when Ms. Luis applied for an apartment in 1990. To get a bachelor apartment, one needed an annual income of $22,000. Whether the criterion is expressed as a ratio or as an annual income determined after the landlord has made calculations, the result is the same.

(b) The Effect of the Factor

113Ms. Luis' complaint cites a number of grounds protected under the Code. The effect of a landlord's use of a rent/income ratio on people in receipt of public assistance is most telling.

(I) RECEIPT OF PUBLIC ASSISTANCE

114Ms. Gravelle testified that in 1990, no one on social assistance would qualify for The Crossways' bachelor apartments. She testified that the "income criteria used at The Crossways would mean that no one on social assistance would qualify for a bachelor apartment". Mr. Stapleton testified that social assistance recipients spend an inordinate amount of their income on rent — the average recipient spends about 50 percent of his/her income. Income criteria tend to exclude people in receipt of social assistance. Mr. Pala testified that if a landlord used a threshold of $22,000 as annual income, no sole-support parent with a child under the age of 9 receiving social assistance benefits would qualify. If landlords used a 30 percent rent/income ratio, over 90 percent of people receiving general welfare assistance would not qualify; over 88 percent of people receiving family benefits assistance would not qualify.

115Dr. Hulchanski's evidence was consistent in pointing out the severe effect of the use of rent/income ratios upon social assistance recipients. The application of rent/income ratios has a significant effect on households receiving public assistance. All households receiving the maximum housing allowance — either under GWA or FBA — would fail a 30 percent rent/income test. All households receiving the maximum housing allowance under FBA would fail a 40 percent rent/income test, while all households receiving a maximum allowance under GWA would fail a 50 percent rent/income test.

116There is ample evidence, including evidence of the respondent's own witness, to support the finding that Creccal's income criteria excluded all recipients of social assistance from The Crossways.

(II) SEX, MARITAL STATUS, FAMILY STATUS, RACE,CITIZENSHIP, PLACE OF ORIGIN

117The evidence was also clear that the use of rent/income ratios by landlords has a disparate impact on a number of other protected groups. Dr. Ornstein's analysis, which included groups categorized by sex, family status, marital status, race and citizenship, noted that in some instances, it is important to recognize that many "groups" intersect and overlap substantially. For example, he emphasized that 90 percent of lone parents are women, so that the effect of a policy on lone parents cannot be isolated from its effect on women. Dr. Ornstein was able to identify general trends from his statistical analysis: unattached people and female lone parents have a high incidence of low income and are affected significantly by the use of rent/income ratios by landlords.

118Evidence of Dr. Leddy was that the source of income for most refugees is welfare, and most refugees pay about 50 percent of their income toward rent.

119The effect of Creccal's policy on Ms. Luis is also important to note. In December 1990, Ms. Luis moved into a different apartment and paid more in rent than she would have had to pay at The Crossways. The apartment was less desirable in terms of its location, its amenities, and its cleanliness.

(c) Membership in the Group(s)

120When Ms. Luis applied at The Crossways, she was clearly a member of the groups identified by race, sex, marital status, family status, citizenship, place of origin, and receipt of public assistance. The evidence shows that Ms. Luis, at the time of applying at The Crossways, was a single black woman with a young daughter, a refugee from Angola, and in receipt of family benefits assistance.

121The Board concludes that all the components of the prima facie case are established against the respondent Creccal. Creccal's practice of using income criterion to screen tenants constitutes prima facie constructive discrimination on the basis of every ground cited in Ms. Luis' complaint.

J.L. v. Shelter Corporation of Canada Limited ("Shelter")

(a) The Existence of the Factor

122J.L. and Marta Dickinson applied for an apartment at Aquitaine Shores, owned by Shelter. The rent was $906 per month. During their application process, they were asked their ages by the rental agent and the application required that they disclose their annual income. Ms. Dickinson testified that two days after delivering the cheques for first and last months' rent:

. . . the property manager told me that we didn't qualify for the apartment, and she gave me back my application . . . and she told me that we weren't making enough money because we both had to be making enough to cover the month's rent . . . and she said specifically that we had to be making over $32,000 a year each.

The application forms were returned, along with the two cheques and written on the face of the rejected application was: "declined — insufficient income + no credit reference June 6/90".

123No evidence was called to contradict Ms. Dickinson's evidence that the figure mentioned by the rental agent was $32,000 per year, and that each applicant had to qualify. An income requirement of $32,000 for an apartment renting at $906 per month amounts to a rent/income ratio of 33.973 percent.

124Further, Shelter's response to the complaint stated: "It is our company's policy that any applicant for tenancy meet our income criteria".

In accordance with Shelter policy, the total amount of rent and utilities to be paid by the applicant should not exceed thirty percent (30 percent).

Mr. Gonsalves, an employee of Shelter, testified that Shelter's rent/income ratio was in the range of 30 percent to 33.33 percent in June 1990. Mr. Gonsalves also testified that in June 1990, Shelter allowed couples to combine their incomes for the purpose of calculating the rent/income ratio, but did not permit two single people to do so.

125The evidence establishes that Shelter was using a rent/income ratio in assessing applications, and that in June 1990, the application of J.L. was rejected, either in whole or in part, because her income level, and that of her co-applicant Ms. Dickenson, did not meet the criterion.

(b) The Effect of the Factor

126There is clear evidence that the impact of rent/income ratios on young unattached women is significant. In Dr. Ornstein's statistical analysis, he noted that 68 percent of unattached women under the age of 20 have less than $10,000 in income. He concluded in his report that "young people are the hardest hit by income criteria ... Young, unattached persons are virtually excluded from accommodation limited by income criteria". Dr. Ornstein concluded further: "Thus, young women are virtually excluded from Aquitaine Shores".

(c) Membership in the Group(s)

127At the time J.L. applied at The Crossways, she was a 16-year-old woman, a member of the groups identified by "age" and by "sex".

128The Board concludes that all the components of the prima facie case are all established against the respondent Shelter. Shelter's practice of using minimum income criterion to screen tenants constitutes prima facie constructive discrimination against J.L. on the basis of age and sex.

Dawn Kearney v. Bramalea Inc. ("Bramalea")

(a) The Existence of the Factor

129The evidence establishes that Bramalea used a rent/income ratio as part of its procedure in assessing applicants for tenancy and did so in assessing Ms. Kearney's application. Bramalea adduced no viva voce evidence as to its practice. However, Bramalea itself identified and referred to the practice of using rent/income ratio in the response to Ms. Kearney's complaint, as follows:

My clients (referring to Bramalea) informed the complainant that in order to occupy an apartment with Bramalea Ltd., any and all applicants must be able to demonstrate an "on-going" ability to pay the monthly rental fees. As such, Bramalea Ltd. had established a policy whereby all applicants must be able to demonstrate that their total income would meet a guideline of 25 percent (or less) of total salary which can be paid towards rent ...

Furthermore, such a criteria is established uniformly and universally for all applicants to Bramalea.

130Ms. Kearney and her husband applied to Bramalea for an apartment in September 1988. Ms. Kearney was pregnant and the couple were seeking a two-bedroom apartment, in the range of $600 monthly rent. The Kearneys told the Bramalea representative their income, and were advised by the Bramalea representative that they had to make $30,000 a year in order to qualify for an apartment.

(b) The Effect of the Factor

131The evidence also supports a finding that the impact of Bramalea's use of a rent/income ratio has a disproportionate effect upon young people. Dr. Ornstein's evidence, particularly the statistical evidence with respect to age and income in the Toronto CMA, and his comparisons between Toronto and Brampton show that the impact was substantial.

132Ms. Callaghan testified as to the effect of landlords' use of rent/income ratios on pregnant and parenting teenagers. She indicated that the most common percentage encountered as a landlord's ratio is 30 percent. She stated that the families that are relying on social assistance, and even those who are working who tend to be in the lower end of paying jobs, are excluded by this ratio.

133The effect of the use of the criterion upon Ms. Kearney was that she and her husband continued to live with her husband's parents in a crowded apartment. There was very little privacy, and tension was increased. The Kearneys, together with one child, shared an extremely small room.

(c) Membership in the Group(s)

134The evidence is clear that Ms. Kearney was 17 years old at the time that she and her husband, who was then 18, searched for an apartment in September 1988. She was pregnant. They were thus both members of the group identified as "young" in the analysis carried out by Dr. Ornstein. Further, as a pregnant woman, Ms. Kearney also came within the group of persons identified by "sex".

135The Board concludes that all the components of a prima facie case are established against the respondent Bramalea. Bramalea's use of [a] rent/income ratio constitutes prima facie constructive discrimination against Ms. Kearney, on the basis of age and sex.

The Effect of the Landlords' Decisions

136In addition to the evidence pertaining to the specific complainants, the Commission submitted that the evidence before the Board also shows that the exclusion of people from accommodation, by landlords' use of rent/income ratios or income criteria, results in the following: people are forced to live farther away from friends and family, from employment, and from schools or child care; the income criteria reduce the stock of apartments for the search; youth wind up in housing that is not as desirable and may be more dangerous, or involve greater problems of hygiene; the more established landlords use the criteria.

137The Board finds that Dr. Ornstein's extensive analysis of the census and other surveys is clear evidence that income criteria differentially affect groups protected by the Code — groups defined on the basis of sex, marital and family status, age, citizenship, race, immigration status, place of origin, and being in receipt of public assistance. The result is to significantly restrict the housing choice of protected groups whose members often end up in higher-priced accommodation of poorer quality.

138The Commission submitted that whether a landlord used a rent/income ratio or a minimum income criterion is immaterial. We agree. Minimum income criteria are set according to the rent of a particular unit, and are merely the landlord's expression of the calculations following the use of a rent/income ratio. In all three cases, the evidence shows the use of a rent/income ratio. In all three cases, the applicants were advised of the minimum income required by application of the rent/income ratio.

139It is well established that if a decision is made for a number of reasons, one of which is prohibited by the Code, then the whole decision is tainted. Proving that one of the factors for a decision is prohibited by the Code is sufficient to prove that a breach of the Code is a proximate cause of the decision. When Shelter rejected the application of J.L. for two reasons — for not meeting the income criteria and having no credit history — and one of the reasons constitutes a prima facie breach of the Code, then the evidence discloses a prima facie breach of the Code.

Section 11 Defence

140Once a prima facie case of constructive or adverse effect discrimination is established by the Commission and the complainants, the respondents must establish a defence on a balance of probabilities. As noted earlier in the decision, it is well established that the interpretation of human rights legislation should be broad and purposive. However, the interpretation of defences, where available, is to be construed strictly. In Zurich Insurance, supra, the Supreme Court of Canada held that [at D/277, para. 58]:

A logical corollary to this purposive approach is the narrow interpretation of exceptions within human rights legislation. Defences to discrimination under the Code must be narrowly construed so that the larger objects of the Code are not frustrated.

141A successful defence under s. 11 of the Code must prove three elements — reasonableness, bona fides, and undue hardship. Section 11 provides as follows:

11(1) A right of a person under Part I is infringed where a requirement, qualification or factor exists that is not discrimination on a prohibited ground but that results in the exclusion, restriction or preference of a group of persons who are identified by a prohibited ground of discrimination and of whom the person is a member, except where,

(a) the requirement, qualification or factor is reasonable and bona fide in the circumstances; or

(b) it is declared in this Act, other than in section 17, that to discriminate because of such ground is not an infringement of a right.

(2) The Commission, the board of inquiry or a court shall not find that a requirement, qualification or factor is reasonable and bona fide in the circumstances unless it is satisfied that the needs of the group of which the person is a member cannot be accommodated without undue hardship on the person responsible for accommodating those needs, considering the cost, outside sources of funding, if any, and health and safety requirements, if any.

142Counsel for the complainants argued that in order for the respondents to prove that using income criteria to screen tenants is reasonable and bona fide, the respondents must meet both a subjective and objective test as required in O'Malley, supra. The respondents must show (1) that the rule was honestly made for sound economic reasons, and out of a genuine belief in its business necessity, and (2) that the rule was equally applicable to all to whom it was intended to apply in order to satisfy the subjective test. The objective test requires that in order for the rule to be found reasonable it, in fact, fulfills a genuine business need and is rationally connected to a genuine business purpose.

143Counsel argued that the evidence was clear that the respondents had adopted and used the rule without any prior evaluation of its ability to measure or predict default. She argued that this was one of the reasons in Griggs (Griggs v. Duke Power Company, 401 U.S. 424 (1971)) that the United States Supreme Court held that a requirement for a high school diploma and a general intelligence test were not related to job performance. The court held:

Both were adopted ... without meaningful study of their relationship to job performance ability. Rather, a vice-president of the company testified, the requirements were instituted on the company's judgment that they generally would improve the overall quality of the workforce.

Thus, counsel argued that the respondents here cannot demonstrate that the rule is rationally connected to a business necessity unless they demonstrate that the income criteria are predictive of tenant default.

144Counsel argued further that under s. 11 of the Code the respondents must also show that individuals adversely affected by the rule cannot be accommodated without undue hardship, "considering cost, outside sources of funding, if any, and health and safety requirements, if any". Counsel argued that when s. 11 was added to Ontario's Code, while it established a similar provision as in the O'Malley decision, it is also clear from the legislative history that it is different. She argued that s. 11 incorporates a more rigorous standard and a more limited list of defences than in O'Malley. She noted that while the O'Malley decision refers to accommodating "the complainant", s. 11 refers to accommodating the "needs of the group of which the person is a member". Counsel argued that the legislative history of s. 11 provides specific factors that can be considered when assessing undue hardship and that it has been held that boards may not consider other factors. Counsel cited Quesnel v. Eidt (1995), unreported [now reported 1995 CanLII 18159 (ON HRT), 28 C.H.R.R. D/474] (Ont. Bd.Inq.).

(a) Reasonableness

145Clause 11(1)(a) of the Code requires that the respondents prove first, on a balance of probabilities, that their use of rent-to-income ratios is reasonable. In order to do this, the landlords must show that there is a rational, objective basis for using such a discriminatory criterion. The landlords must prove that there is a rational connection between their use of a rent-to-income ratio, and the landlords' business objective of minimizing instances of default. That is, the landlords must show that their use of rent-to-income ratios reduces the landlords' incidence of default.

146Having carefully considered the evidence of the respondents, we find that there is no evidence to show that the use of a rent-to-income ratio reduces a landlord's incidence of default. On the contrary, there is considerable reliable evidence that shows that there is no relationship between a tenant's rent-to-income ratio at the time of entry into the landlord/tenant relationship, and the probability of default by that tenant at some future point. In the absence of evidence connecting rent-to-income ratios with default, the Board does not have to consider whether the respondents have satisfied the burden of proof. It is not an issue.

147The history of the "rule" shows that it is not based upon empirical data. There is no relationship between the percentage of income that a person spends on rent and the probability that such person will default in the payment of rental obligations. The rule has its roots in the nineteenth century studies of how households budgeted their resources and in the commonly used expression "one week's pay for one month's rent". By the late nineteenth century, a week's wage for a month's rent was widely used for referring to the housing expenditures of some tenants. This continued into the twentieth century, changing upward to 20 percent, 25 percent or 30 percent as somehow representing the upper limit of what households could afford to pay for housing. We are persuaded by the evidence of Dr. Hulchanski (summarized more fully above) that the rule is based on little more than assumptions about what average households tend to spend and/or beliefs about what they ought to spend on housing.

148Dr. Hulchanski's evidence identified six contemporary uses of the rent-to-income ratio, and classified them as (1) description, (2) analysis, (3) administration, (4) definition, (5) prediction, and (6) selection. The latter three uses, defining "need" for policy purposes, using rent-to-income ratios to predict whether tenants will default, and using rent-to-income ratios for selection of tenants, are all improper. The use of the income criteria for these purposes leads to invalid and unreliable results. Using income criteria for defining need, predicting default, and selecting tenants is invalid because it fails to measure what it claims to be measuring, even if the research methods and the statistical analysis techniques are properly carried out.

149Using income criteria to predict default is specifically analyzed and rejected as invalid by Dr. Hulchanski. It was his opinion that there is no evidence that the ratio of housing expenditure-to-income is a valid and reliable measure of ability to pay rent. Since households may respond differently, and yet effectively, to changing circumstances, imposing the same standard for all households is unrealistic. A maximum rent-to-income ratio for one household may not be appropriate for another.

150Dr. Hulchanski's evidence was that households meet their needs through a variety of methods. The rent-to-income ratio, by using only formal money income, fails to recognize that households use far more than the formal economy in meeting needs; the ratio ignores other sources of support. The inadequacy of the measure of "income" used in the income criterion is itself a ground for rejecting the use of the criterion. Further, there is no evidence to support its use as a measure of ability to pay for housing. There is substantial evidence to the contrary — evidence that many households are paying much more than the prescribed ratio. The reality of how households manage to meet their needs, including the need to have the cash to pay their rent, is too complex and diverse to be summarized in one simple measure.

151Also, the evidence relating to the rental market shows that the rent-to-income ratio is not a predictor of the probability of default. For example, Dr. Hulchanski points out in his Background Paper number 2 that in Ontario (1991 data), some 830,000 renters and owners, (about 23 percent of Ontario households), should be defaulting on their mortgage and rent payments if the ratio of 30 percent were a valid predictor of default. There is simply no evidence that default rates come anywhere close to those levels.

152Further, it is clear from the evidence presented by Dr. Hulchanski that a landlord's use of a rent-to-income ratio in tenant selection does not help predict (a) whether a tenant will pay rent as it becomes due, (b) whether a tenant will be able to pay rent as it becomes due, and (c) whether a tenant will be willing to pay rent as it becomes due. The academic literature cited by Dr. Hulchanski on the topic of the use of a rent-to-income ratio in housing is that use of a ratio has no value in predicting the likelihood of default.

153The evidence supports a conclusion that it is unexpected changes in one's circumstances after entering into a tenancy which are the most common cause of a tenant's default.

154Most importantly, the evidence establishes that the practices of the landlords in using income criteria is not reasonable. Bramalea admits in the response to the complaint of Ms. Kearney that it used a 25 percent rent-to-income ratio. However, no evidence was called as to why Bramalea used a rent-to-income ratio, or why a particular ratio was adopted. Thus, the only possible conclusion for this Board is that there is "no evidence that Bramalea used a specific rent-to-income ratio in its tenant selection process because the 25 percent ratio had any predictive value in indicating whether a tenant would default in the payment of rental obligations as they become due".

155Testifying for Shelter, Mr. Gonsalves indicated that Shelter used a rent-to-income ratio of "one-third" and Shelter's written policies show that the rent-to-income ratio to be applied was 30 percent. However, Mr. Gonsalves gave no evidence that Shelter used a rent-to-income ratio because it had any predictive value in determining whether a tenant would pay rent in a timely way or because it had predictive value in indicating whether a tenant would default.

156Creccal's evidence as to the use of a rent-to-income ratio came from Mr. Di Geso who said Creccal used the industry standard of 30 percent as an income criterion. The evidence shows that Creccal first used an income criterion of $22,000, annual income requirement, based upon an annual rent of $6,000 (a rent-to-income ratio of 27 percent), and continued to use the income criterion of $22,000 even when average rents had risen to $7,800 (a rent-to-income ratio of 34 percent). The evidence was that over a period of a few years, Creccal moved its rent-to-income ratio from 27 percent to 34 percent and did so with no effect upon its default rate.

157Mr. Di Geso also indicated in evidence that his application of the income criterion changed for tenants in receipt of public assistance. For tenants in receipt of public assistance he would agree to the tenancy provided that the amount of rent did not exceed the person's shelter allowance as set out in the rates for public assistance benefits. Thus he would rent an apartment to a single mother, in receipt of public assistance, even though the rent-to-income ratio would exceed 50 percent of benefits. Mr. Di Geso testified that he would be able to rent an apartment since "it's guaranteed she's getting that $585 to pay as rent". There was no evidence from Mr. Di Geso that by agreeing to take tenants whose rent would exceed 50 percent of income there was an effect upon the default rates in Creccal's apartments.

158The evidence of the three landlords shows that they used various rent-to-income ratios — 25 percent, 27 percent, 30 percent, 33 1/3 percent, 34 percent. However, there is no empirical evidence that any of the ratios was picked because it had any predictive value with respect to the issue of whether a tenant would default in the payment of rent. The landlords were simply using "numbers" which had their historic bases as clearly outlined in the evidence of Dr. Hulchanski. The landlords were not using rent-to-income ratios because there was an objective rational basis for so doing.

159The evidence tendered by the respondents fails to show a connection between a rent-to-income ratio and the risk of default. The respondents did not submit empirical evidence comparing the default experiences of landlords who use rent-to-income ratios and those who do not. The respondents' own witnesses, Professors Trebilcock and Halpern, stated that they advised the respondents that empirical work was needed; in fact, they insisted that it be done. Professors Trebilcock and Halpern indicated that they themselves had carried out no empirical studies comparing the default experiences of landlords who use rent-to-income ratios and those who do not. But their opinions rested on the premise that income criteria were predictive of default.

160The only empirical evidence tendered by the landlords, in an attempt to show that the use of a rent-to-income ratio by a landlord is "reasonable", comes in the form of an analysis carried out by Mr. Earwaker. Mr. Penner's report relied on this statistical analysis. Professors Trebilcock and Halpern also relied upon Mr. Earwaker's analysis for the proposition that the use of a rent-to-income ratio reduces the risk of default. Similarly, Mr. Lampert, who was retained to give evidence on behalf of the landlords, acknowledged that he did not do a survey of landlords who use rent-to-income ratios in comparison to those who do not, and relied on Mr. Earwaker's and Mr. Penner's analysis.

161Mr. Earwaker, a statistician, was approached by Mr. Penner to analyze data from the two respondents, Bramalea and Shelter. Mr. Earwaker testified that he compared three groups — defaulting tenants, delinquent tenants, and non-delinquent/non-faulting tenants. Mr. Earwaker's analysis compares the average rent-to-income ratios of those three groups. The question posed by Mr. Earwaker was: what are the average rent-to-income ratios of the three groups — defaulting tenants, delinquent tenants, and non-defaulting/ non-delinquent tenants?

162However, the question as posed by Mr. Earwaker has no bearing on the essential issue to be addressed in this case: does using a rent-to-income ratio in tenant screening and selection reduce a landlord's risk of default? Mr. Earwaker acknowledged that he did not test the hypothesis that a landlord's use of a 30 percent rent-to-income ratio will reduce that landlord's default experience. Mr. Earwaker admitted in cross-examination that that question would require "a different test". Mr. Earwaker also acknowledged that there is no direct translation between what he did — analyzing average rent-to-income ratios — and analyzing the risk to landlords.

163The evidence of Dr. Ornstein, given in reply, explains the fundamental error in Mr. Earwaker's formulation of his expert report:

To conduct analysis bearing directly on the efficacy of income criteria, it is only necessary to compare the default rates of tenants with rent-to-income ratios above and below the specific income criteria employed by the respondents. This Mr. Earwaker did not do. Instead, he chose to compare the average rent-to-income ratios of the non-defaulting non-delinquent, delinquent and defaulting tenants and as noted above, did not do so correctly.

It seems likely that if defaulters have a higher mean rent-to-income ratio than non-defaulters, then landlords' use of a specific criterion will lower the number of defaults. Differences in the average rent-to-income ratios of the three groups, however, cannot be translated into a test of the efficacy of any particular income cut-off in the determination of the eligibility for accommodation. We cannot assume that the effects of different criteria are "linear". For example, showing that a 35 percent criterion decreases the likelihood of default would not necessarily mean that a 30 percent criterion would decrease the likelihood still more and a 25 percent criterion even further. Mr. Earwaker could have designed his study to analyse the impact of the specific income criteria used by the respondents, but did not.

Dr. Ornstein emphasized in his evidence in reply that the design of a relevant study should reflect the research question. However, Mr. Earwaker's does not and therefore his analysis does not support a finding that the rent-to-income ratio helps predict default. Dr. Ornstein emphasized that in Mr. Earwaker's work, there is no analysis of whether the use of any particular ratio affects the risk of default. There were also critical errors in the methodology.

164The Board agrees with Dr. Ornstein's opinion that Mr. Earwaker's research was fundamentally flawed, and the analysis does not address the question of the efficacy of rent-to-income rations as predictors of default. It is, therefore, of no assistance to the Board.

165It is the Board's conclusion that the practice of using a rent-to-income ratio is reasonable only if the use of such a ratio reduces a landlord's risk that tenants will default in the payment of rental obligations. The evidence presented by the respondents fails to show any connection between landlords' use of rent-to-income ratios and the reduction of the risk of default. Therefore, the landlords have failed to show that their use of rent-to-income ratios is related to a genuine business need. Consequently, the respondents have failed to prove a necessary component of the defence required under clause 11(1)(a) of the Code.

(b) Bona Fides

166The respondents also failed to establish with evidence that the use of rent-to-income criteria is bona fides, an element of a defence under clause 11(1)(a). The jurisprudence on bona fides is found in mandatory retirement cases such as Etobicoke, supra, and Large v. Stratford [(1995), 1995 CanLII 73 (SCC), 24 C.H.R.R. D/1]. In those cases, the Supreme Court held that a bona fide qualification had both a subjective and an objective component. "Reasonableness" was not included in the statutory provision considered by the courts, since the predecessor of clause 24(1)(b) considered in these cases, dealt only with whether the requirement was bona fide. The objective component from the earlier cases is reflected in the incorporation of "reasonableness" into clause 11(1)(a). Thus, the required component of bona fides, as it appears in s. 11, can be analyzed from the jurisprudence on the subjective component of bona fides in those earlier cases. In Etobicoke, McIntyre J. defined the subjective arm of the test as follows [at D/783, para. 6894]:

To be a bona fide occupational qualification and requirement a limitation, such as mandatory retirement at a fixed age, must be imposed honestly, in good faith, and in the sincerely held belief that such limitation is imposed in the interests of the adequate performance of the work involved.

167In Large v. Stratford, the Court had the opportunity to consider the test set out in Etobicoke, in circumstances in which it was clear that the employer had imposed the mandatory retirement age at the request of the union. Sopinka J. observed as follows (Large v. Stratford (City)(1995), 1995 CanLII 73 (SCC), 24 C.H.R.R. D/1 at D/8, para. 21):

... The purpose of the subjective element is to ensure that a discriminatory rule was adopted for a valid reason, as an occupational requirement, and not for a prohibited, discriminatory reason. It ensures that the employer is not attempting to defeat the purposes of the Code, in short, that the employer does not have a discriminatory motive. Usually, this goal will be realized and the subjective element established by evidence that the employer honestly believed that the qualification or requirement was necessary for the safe and/or effective carrying out of the work.

Sopinka J. went on to explain that it was wrong for the tribunal and the lower courts to insist on evidence as to the employer's state of mind [at p. D/9, para. 22]:

... In some circumstances the subjective element can be satisfied when, in addition to satisfying the objective test, the employer establishes that the rule or policy was adopted in good faith for a valid reason and without any ulterior purpose that would be contrary to the purposes of the Code.

Thus, the respondents must show on a balance of probabilities that they used rent-to-income criteria in the honest belief that it was necessary for their businesses.

168The Board finds that the respondents have failed to establish the bona fide component of the defence. There was clear evidence that Shelter and Creccal did not always apply their rent-to-income criteria. No evidence at all was adduced as to why Bramalea used a ratio. Mr. Gonsalves for Shelter used a ratio simply because it had come from head office; Mr. Gonsalves clearly did not view the use of a rent-to-income ratio as necessary — he tended to treat it as directory as opposed to mandatory. Mr. Di Geso, for Creccal, was clearly prepared to change the ratio he used. He does not view the use of a ratio as necessary. In accepting tenants for whom the rent did not exceed the shelter component of public assistance benefits, Mr. Di Geso was not using a ratio at all.

169Further, given the evidence that a substantial number of landlords do not use rent-to-income ratios, and that some landlords appear to use the criteria selectively, the Board cannot conclude that landlords have an honest belief that applying rent-to-income criteria to applicants is necessary for the effective carrying out of the business. Given the onus of proof, this Board finds that the respondents have not proven bona fides.

170The respondents have failed to establish the second requirement of a successful defence under clause 11(1)(a) of the Code; that their use of rent-to-income ratios was bona fides.

(c) Undue Hardship

171Section 11 of the Code requires that the respondents prove on a balance of probabilities that they will suffer "undue hardship" if they are required to stop using rent-to-income ratios in their screening of tenants. The jurisprudence has established that in considering what counts as undue hardship, some hardship is "due". As the Supreme Court held in Central Okanagan School District No. 23 v. Renaud(1992), 1992 CanLII 81 (SCC), 95 D.L.R. (4th) 577 [16 C.H.R.R. D/425 at D/432, para. 19] (S.C.C.): "The use of the term ”˜undue' infers that some hardship is acceptable; it is only ”˜undue' hardship that satisfies this test." Thus, it is not enough for a respondent to show that there will be a cost; the cost must be shown to be excessive or disproportionate. The cost must be of such significance that a board of inquiry would permit a practice that is prima facie discriminatory. The onus is on the respondents to prove undue hardship on a balance of probabilities.

172Commission counsel argued that hardship if it is to be proven by the respondents, it must be balanced against the fundamental goals of the Code. Counsel cited Janssen, supra, where the Tribunal balanced the cost of accommodation against the importance of freedom of religion, and Emerson Electric, supra, where the Board referred with approval to the balancing of undue hardship against the goals of the Code to protect against discrimination.

173Thus, counsel argued that, in the cases before this Board, any hardship, if proven, would have to be balanced against the following:

(i) recognition of the dignity and worth of individuals;

(ii) the creation of a climate of understanding and mutual respect;

(iii) the goal that each person feels a part of the community; and

(iv) effective recognition of those goals of the Code (reflected in its preamble), particularly in light of the fundamental need of housing and the specific inclusion in the Code of provisions with respect to occupancy of accommodation such as s. 4 (dealing with people aged 16 and 17) and s. 2 (dealing with the receipt of public assistance).

174We have decided that since the rent-to-income ratio has no predictive value as to whether a tenant is likely to default in future, stopping the use of such a ratio will not cause hardship. The evidence of Dr. Hulchanski noted as follows:

Not using an erroneous, invalid measure as part of the criteria in the decision to rent to a particular household means, of course, that landlords will either be unaffected or that they will be better off. Landlords will be unaffected because the measure they are using to assess the ability to pay does not actually measure the ability to pay. Nothing changes when an erroneous measure ceases to be used. Landlords will profit by not using this (or any) invalid measure in that, if they have vacancies which they delayed filling due to rejection of tenants based on the use of such a measure, they will fill these vacancies more quickly.

For essentially the same reasons outlined above — that the landlords have not proven that the use of rent-to-income ratios is reasonable — the landlords have failed to prove that ceasing the use of those ratios will cause them any hardship. The respondents have adduced no reliable evidence that the rate of default would be any different if landlords were to stop using rent-to-income ratios in tenant selection. Since the respondents have failed to show that discontinuing the use of rent-to-income ratios in tenant selection will increase the rate of default (or delinquency) by tenants, they have clearly failed to show that any hardship will be caused. Therefore, the landlords have failed to prove that undue hardship will result from discontinuing the use of income criteria.

175A great deal of evidence was adduced before this Board as to the average costs of a default or a delinquency. Since all of that evidence relates to the cost of a default or delinquency, none of it is relevant to the fundamental question before the Board, which is whether using a rent-to-income ratio to screen tenants reduces the landlord's risk of default and is, therefore, a reasonable bona fide business requirement.

176In summary, the Board finds that the respondents have failed to establish the components of a defence under s. 11 of the Code. It is clear in the evidence that landlords do not apply the income criteria consistently, and that there is no evidence to support a conclusion that there is a greater risk of default when landlords do not use income criteria. They have failed to tender evidence that income criteria is rationally connected to a business necessity.

Submission on Direct Discrimination

177Complainants' counsel made an extensive argument that the Board should find that the landlord's use of rent-to-income criteria to screen tenants is direct discrimination.

178Counsel argued that the use of income criteria was inherently discriminatory and, therefore, direct discrimination. Counsel submitted that one of the considerations in determining whether a policy or rule was inherently discriminatory was the extent of its exclusionary impact. If the impact experienced is more than a small minority, individual accommodation becomes inappropriate and the only remedy is to strike down the policy.

179Counsel argued that a rule may constitute adverse effect discrimination against one group and direct discrimination against another one. A rule may be directly discriminatory against a protected group who are completely, or almost completely, excluded by its application. The same rule may have an adverse impact upon another protected group who experience a lesser, but still statistically significant discriminatory effect as a result of the rule. Counsel argued that the only remedy is to strike the policy in its universal application. When multiple groups experience a discriminatory effect of a policy, the court prefers to strike down the policy rather than to maintain it and accommodate those adversely effected. Counsel cited Central Alberta Dairy Pool v. Alberta (Human Rights Commission)(1990), 1990 CanLII 76 (SCC), 72 D.L.R. (4th) 417 [12 C.H.R.R. D/417] (S.C.C.).

180Counsel argued that a prima facie case of direct discrimination is established where it is proved that:

(a) The impugned policy or practice had a discriminatory effect or imposes a burden upon the group identified by a prohibited ground of discrimination, and

(b) The impugned policy or practice is properly characterized as direct discrimination on the basis of one or more of the following indicia:

(i) The policy is facially discriminatory: A policy is facially discriminatory if it differentiates on the basis of membership in a protected group, or on the basis of a characteristic closely associated with the group.

(ii) The policy is intentionally discriminatory: A policy is intentional if the discriminatory effect of the policy on the identified group was intended by the respondent, known by the respondent or reasonably foreseeable by the respondent.

(iii) The policy is extensive in its effects. A policy is directly discriminatory if the extent of the discriminatory effect on members of protected group is large and significant.

(iv) The policy is inherently discriminatory so that the appropriate remedial response is to strike it down. In cases of direct discrimination the extent of the exclusionary impact is such that individual accommodation is untenable. Consequently, the appropriate remedial response is to strike the rule down.

181Since we have decided that a landlord's use of income criteria to screen tenants results in adverse effect or constructive discrimination against the protected groups identified in the complaints, and that the use of income criteria is neither reasonable, bona fide, or would cause undue hardship if it was not permitted, we have decided not to address the argument that the use of income criteria to screen tenants amounts to direct discrimination. We make this decision in part because we have decided that the use of income criteria has an impact which is so pervasive that it is proper to declare that the rule is contrary to the Code, and strike it in its general application.

182Section 11 of the Code does not refer to a distinction made in O'Malley and subsequent cases by the Supreme Court that in most, but not in all cases, where adverse effect discrimination is found, the appropriate remedy is to uphold the rule in its general application, but disallow its use with regard to individuals who have suffered a disparate impact by its application. In cases of adverse effect discrimination where the impact is so pervasive, it is proper to declare the rule as contrary to the Code. In Griggs, supra, referred to by the majority in O'Malley, supra, the United States Supreme Court struck down the requirement of a high school diploma and intelligence tests because of their extensive adverse impact on African Americans. In Colfer v. Ottawa Board of Commissioners (1979), unreported (Ont. Bd.Inq.), the Board struck down a height and weight requirement because of its adverse impact upon women applying to be part of the police force. We are of the view that income criteria belongs to the kind of adverse effect cases where the impact of the rule is so pervasive that the remedy must be to strike down the rule in its general application.

Respondents' Submission

183Before addressing the question of what remedy is appropriate in these cases, we should comment briefly on the respondents' submission to the Board, made by Mr. Doane for Bramalea and Shelter, and adopted by Mr. Luftspring for Creccal.

184Mr. Doane argued that the case before us involves the way a scarce good — rental accommodation — is sold or leased in the Ontario market. Professors Halpern and Trebilcock testified for the respondents that the rental contract is a form of credit contract: the landlord lends its asset to a tenant — an apartment unit — in exchange for the tenant's promise to make certain periodic payments until possession of the asset is returned. Like all lenders, the landlord faces a financial risk that the payments may not be made. This risk includes not only loss of the contracted payments, but also property damage and various eviction and re-leasing costs. Counsel asked us to look at the evidence presented by the respondents which indicated that in various ways a broken rental contract has significant costs to a landlord that can reach many thousands of dollars. Counsel argued that for small landlords, a single broken contract can mean financial ruin.

185Counsel submitted that prudent lenders employ a pre-contract process for limiting the risk of broken contracts and resulting costs. The rental contract, though, has two important distinctions from any other commercial credit contracts. The first distinction is that with a commercial credit contract, a lender can manage risk through manipulation of the price system (interest rates). Landlords are prevented from charging differential rents based on risk. The second distinction is the possibility of taking security up to the full value of the contract to further reduce the risk of default. This option is not available to landlords. The absence of these risk management mechanisms leaves landlords in a comparatively precarious position, and thus even more strongly encourages pre-contract screening.

186The Board recognizes that landlords are in the business of "selling" a service and they have a right to use criteria to assess risk of loss. However, landlords cannot use a criteri[on] that adversely impacts protected groups under the Code when the criteri[on] has not been shown to be reasonable or bona fide.

187Counsel noted that a landlord cannot perfectly predict an applicant's ability to pay or estimate expected default costs. As Professors Halpern and Trebilcock stressed in their evidence, landlords, like other credit grantors, are always dealing with future probabilities rather than with certainties. However, not only is the information that is available to landlords an imperfect predictor, it is also incomplete: the need for "perfect" information and the benefits of more information must be balanced against the costs of acquiring it. Thus, "in the face of the need to manage credit risk, while at the same time confronted with the reality of imperfect, costly information, landlords are forced to use proxies such as income criteria to assess future ability to pay and other risk factors". Income criteria measure the economic burden that rental obligations would pose on tenants, and assist in "determining the stability of income and the level of financial resources that could be accessed upon default". Counsel argued that income criteria "may not be perfect, but it's all we've got".

188As noted earlier, we recognize that the landlords have a valid interest in managing risk. However, the landlords presented no evidence to us that use of income criteria reduced the incidents of default or was effective in assessing future ability to pay. It is not enough to argue that it's "all we've got".

189Counsel argued further that, to the respondent's knowledge, no private rental accommodation market in Canada or elsewhere in the industrialized world has a form of random allocation of housing. Counsel argued that by preventing the use of income criteria by landlords, all prospective tenants would have to be treated as equally meritorious (when clearly this is not true). Further, counsel submitted that:

... a random allocation regime is likely to impose significant additional costs on landlords, who would henceforth be foreclosed from assigning any weight to differential default risk in allocating apartments. The consequences are analogous to the costs entailed in requiring automobile suppliers to sell cars below cost to buyers without the cash to pay for them, or analogous to the case of suppliers of goods on credit who are required to supply the goods on credit, or banks being required to provide mortgage financing, irrespective of the ability of the consumer to meet future credit commitments, with likely future implications in terms of disruption in this market through reduced supply of rental accommodation and higher costs for all tenants.

Finally, counsel argued a random allocation regime would involve cross-subsidization by low-risk tenants (individually meritorious tenants) of high-risk (less meritorious) tenants of the kind that Sopinka J., speaking for the majority in Zurich, supra, viewed as an objection to common pooling of widely disparate insurance risks.

190This Board was asked to consider whether the use of income criteria in assessing tenants for rental housing violated the Code. The Commission and the complainants proved in their evidence that income criteria do not predict default. They also showed that the use of income criteria has a substantial adverse impact on groups protected by the Code. The Commission and complainants at no time asked this Board to find that random allocation of rental housing units is appropriate. Landlords are entitled to seek credit history, rental history, employment status, and use other selection criteria that do not violate the Code. Landlords, however, must be mindful that there is a difference between a tenant having no credit rating and a bad credit rating. There is a difference between a poor reference from a previous landlord and no reference. Denying housing to a prospective tenant because the individual has no credit rating, no landlord or employment history, is like denying housing for failing to meet an income criteria in our view.

191Counsel argued that the Code does not impose wide-scale affirmative action programs. The objectives of such programs are fundamentally different than the Code's objective to protect individuals so that they are evaluated on their merit. Wide-scale affirmative action programs — programs used to alter aspects of the market (e.g., pay equity) — are carefully crafted programs with redistributive objectives whose conceptual and legal companions are other supply-side programs designed to improve the life prospects of members of historically disadvantaged groups.

192Counsel argued that:

The theory of the case presented by the Commission and the complainants, when carefully analyzed, assigns a fundamentally redistributive role to human rights legislation in reducing economic inequalities, and seeks to radically reconceive the objectives of this legislation in a way that is at variance with its normative precepts and that is likely to generate long-term empirical consequences that are in fact adverse to the various groups on whose behalf this case is being advanced.

Counsel made an extensive submission on market economics and economic inequalities to support his argument. For the same reasons as noted above, the Board finds this argument unhelpful. The Commission is not suggesting that this Board order a "wide scale affirmative action" program. The point is there is no evidence that the use of income criteria does what some landlords may well think it does — predict default. There is ample evidence from the experts and from academic research, cited by the experts, that it does not predict whether an individual will default on paying rent. Therefore, there is no reason to believe that landlords will be affected by not being able to use the income criteria they have been using. Moreover, the complainants were simply seeking the chance to pay market rent. They were not asking for special rates or special treatment.

193Finally, counsel submitted that the Code's focus on equality of opportunity is grounded textually, both in its preamble and in its restricted definition of "equal". The preamble provides in part:

AND WHEREAS it is public policy in Ontario to recognize the dignity and worth of every person and to provide for equal rights and opportunities without discrimination that is contrary to law, and having as its aim the creation of a climate of understanding and mutual respect for the dignity and worth of each person so that each person feels a part of the community and able to contribute fully to development and well-being of the community and the Province.

The definition of "equal" in s. 10 states that "”˜equal' means subject to all requirements, qualifications and considerations that are not a prohibited ground of discrimination". Counsel argued that this definition reflects the concept of equality and a tolerance for unequal results which occur because of differences in social and material conditions, individual talents, and choices.

194Counsel's argument here fails to appreciate that an otherwise neutral rule which adversely impacts groups protected by the Code will be found to be adverse effect discrimination unless the respondent can show that the rule is reasonable and bona fide and further that it would not cause the respondent undue hardship to accommodate the complainant. In an early ruling on this issue, in Griggs, supra, the United States Supreme Court found that the requirement of a high school diploma and intelligence tests used to screen prospective employees, which virtually screened out all African Americans, was discrimination. The rule was simply unreasonable. There was no evidence that the high school diploma and tests were rationally connected to a business need. Therefore, the unequal results of these employment requirements were not tolerated. Unequal results are permitted under the Code if the rule or criteria used to discriminate or distinguish candidates is reasonable and bona fide and the complainant cannot be accommodated without undue hardship.

SUMMARY

195After careful consideration of the submissions of the parties and the interveners (a summary of the submissions of the interveners is at Appendix 1 [not published here]) and review of the extensive evidence presented to the Board, we have decided that the respondents' use of income criteria to exclude the complainants from housing in their respective buildings constitutes adverse effect or constructive discrimination. The Commission and complainants' evidence established a prima facie case for each complainant. The evidence was clear that the use of income criteria is not a valid predictor of default. There was substantial evidence that the use of the criteria disproportionately excludes groups protected by the Code from rental housing. Once the Commission had proven a prima facie case, the respondents had to prove that the use of the income criteria was reasonable and bona fide and results in undue hardship to the respondents if they were required to accommodate the complainants. The respondents failed to put forth evidence before the Board to discharge this onus. There was simply no evidence to support a finding that the use of income criteria in the selection of tenants is reasonable and bona fide. We find that the use of income criteria to select tenants violates the Code, whether it is used by itself or in conjunction with other selection criteria.

196We are of the view that an order declaring that the requirement of a guarantor/co-signor, credit rating, employment or landlord history is beyond the scope of the complaints before us, which clearly challenged the use of income criteria as contrary to the Code. We make this decision after carefully considering our broad remedial powers under s. 41 of the Code. As noted earlier, we are of the view that landlords should be mindful of the difference between a positive or negative credit history and no credit history, a landlord history and no history.

197In summary we find that the evidence clearly shows that:

(i) in the case of Catarina Luis and her application to Creccal, Creccal used an income criterion or rent/income ratio, amounting to constructive discrimination, and thus is in breach of ss. 2 and 9 of the Code;

(ii) in the case of J.L. and her application to Shelter, Shelter used an income criterion or rent/income ratio, amounting to constructive discrimination, and thus is in breach of ss. 2, 4 and 9 of the Code; and

(iii) in the case of Dawn Kearney and her application to Bramalea, Bramalea used an income criterion or rent/income ratio amounting to constructive discrimination, and thus is in breach of ss. 2, 4, and 9 of the Code.

ORDER

198Having found that a landlord's use of a rent/income ratio, or of an income criterion — whether alone or in combination with other assessment factors — in assessing applications for residential tenancy, breaches the Code, and having found that the rights of the three complainants were breached by the actions of the respective respondents herein, the Board makes the following orders:

1. The Board declares that the use of rent-to-income ratios/minimum income criteria violate ss. 2 (1), 4, 9 and 11 of the Human Rights Code whether used alone or in conjunction with other selection criteria or requirements.

2. The respondents, Creccal Investments, Shelter Corporation and Bramalea, shall cease and desist from using rent-to-income ratios or minimum income criteria in selecting prospective tenants whether alone or in conjunction with other selection criteria or requirements.

3. The respondent, Creccal, shall pay Ms. Luis specific damages of $460 for losses arising when she was denied housing at The Crossways.

4. The respondent, Creccal, shall pay Ms. Luis $5,000 as general damages for its breach of ss. 2(1), 9 and 11 of the Code.

5. The respondent, Creccal, shall pay Ms. Luis pre-judgment interest on the amounts set out in 3 and 4 above from May 4, 1992, calculated in accordance with the Courts of Justice Act, R.S.O. 1990, c. C.43, s. 27 rate.

6. The respondent, Shelter, shall pay J.L. $4,000 as general damages for its breach of ss. 2(1), 4, 9 and 11 of the Code, and pre-judgment interest on that amount from June 25, 1990, calculated in accordance with the Courts of Justice Act rate.

7. The respondent, Bramalea, shall pay Ms. Kearney $4,000 as general damages for its breach of ss. 2(1), 4, 9 and 11 of the Code and pre-judgment interest on that amount from September 23, 1988, calculated in accordance with the Courts of Justice Act rate.

8. Post-judgment interest pursuant to the Courts of Justice Act shall begin to accrue on all damage awards one month from the date of this decision.

199The Board shall remain seized of the case to deal should there be any difficulties in the implementation of the orders.