CITATION: Fulton v. Guan et al, 2026 ONSC 2757
DIVISIONAL COURT FILE NO.: DC-25-00000437-00JR
DATE: 20260807
ONTARIO
SUPERIOR COURT OF JUSTICE
DIVISIONAL COURT
Matheson, Muszynski, and Brownstone JJ.
BETWEEN:
Benjamin Fulton
Applicant
– and –
Hemei Guan, Xiaojun Tao, Ontario Human Rights Commission and Human Rights Tribunal of Ontario
Respondents
Shibil Siddiqi, for the Applicant
Hemei Guan and Xiaojun Tao, Self-Represented
Reema Khawja and Alisha Krishna, Counsel for the Respondent Ontario Human Rights Commission
Hera Evans and Elizabeth Guilbault for the Attorney General of Ontario
Mindy Noble, Counsel for the Human Rights Tribunal of Ontario
HEARD at Toronto: February 23, 2026
REASONS FOR DECISION
L. BROWNSTONE J.
Overview
[1] The applicant, Benjamin Fulton, brings this application for judicial review of the decision of the Human Rights Tribunal of Ontario dated May 1, 2025, dismissing his application.
[2] Mr. Fulton is blind. At the time of the events that led to his human rights application, he was a law student at Osgoode Hall Law School. Mr. Fulton has had difficulty securing affordable housing. He sought shared accommodation in owner-occupied housing.
[3] Upon learning he was blind, the respondent Hemei Guan denied Mr. Fulton a room in the home she owns with the respondent Xiaojun Tao, advising Mr. Fulton that the room was not suitable for him. Ms. Guan said she was concerned Mr. Fulton would have an accident.
[4] The Human Rights Code R.S.O. 1990, c. H.19 prohibits discrimination in accommodation on various protected grounds. Disability is one such ground. However, s. 21(1) of the Code exempts housing in which tenants share a kitchen or bathroom with the owner or the owner’s family from the Code’s protection against discrimination.
[5] Mr. Fulton applied to the HRTO, asserting that he had been discriminated against on the basis of his disability. Within his application, he challenged s. 21(1) of the Code, submitting that it violated his right to life, liberty and security of the person under s. 7 of the Charter and his rights to equal protection and benefit of the law under s. 15(1) of the Charter. He asked the HRTO to refuse to apply s. 21(1) to his application on constitutional grounds.
[6] The HRTO dismissed Mr. Fulton’s application. In addition to finding there was no s. 7 violation, the tribunal held that Mr. Fulton had not met the first part of the test under s. 15(1). That branch of the test required Mr. Fulton to establish that, on its face or in its impact, the impugned law creates a distinction based on enumerated or analogous grounds. Having determined that Mr. Fulton failed to meet the first branch of the s. 15(1) test, the HRTO did not consider the second part of the test, or the arguments made under s. 1 of the Charter.
[7] Mr. Fulton does not pursue his s. 7 claim in this application.
[8] Mr. Fulton submits that the tribunal erred in finding that he did not meet the first part of the s. 15(1) test. He asks the court to overturn the finding, find that he met the first part of the test, and engage in the remaining steps of the Charter analysis.
[9] For the reasons that follow, I find that the HRTO erred in its analysis under s. 15(1). I find that Mr. Fulton met the first part of the s. 15(1) test. On this basis, I would allow the application. However, rather than undertaking the remainder of the Charter analysis as Mr. Fulton requests, I would remit the matter to the HRTO to undertake that exercise.
Issues
[10] This application raises the following two issues:
Was the tribunal correct in determining that the applicant’s evidence did not support a finding or inference that s. 21(1) of the Code has a disproportionate impact on individuals with disabilities, and therefore did not satisfy the first step of the s. 15(1) analysis?
If the tribunal was incorrect, what is the appropriate remedy?
Standard of review
[11] The applicant and the Attorney General submit that, because the case raises a constitutional issue, a correctness standard of review applies: Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, at paras. 17, 53 and 55. The applicant and the Attorney General submit that this standard of review extends to findings of mixed fact and law made in the context of analyzing a constitutional question: Société des casinos du Québec inc. v. Association des cadres de la Société des casinos du Québec, 2024 SCC 13, 491 D.L.R. (4th) 385, at para. 45.
[12] The HRTO submits that “[w]here it is possible to treat the constitutional analysis separately from the factual findings that underlie it, those factual findings are accorded deference and assessed on a standard of reasonableness.” I agree that Société des casinos du Québec inc., at paras. 96-97 holds that extricable findings of fact are entitled to deference. However, in this case, the findings at issue are findings of mixed fact and law, that is, “those that determine ‘whether the facts satisfy the applicable legal tests’”: Société des casinos du Québec inc., at para. 94 (internal citation omitted).
[13] I therefore agree with the applicant and the Attorney General that the applicable standard of review is correctness.
The Applicant
[14] Mr. Fulton is a lawyer who was called to the bar in 2019. He is blind and has had difficulty finding and maintaining employment.
[15] At the time of the hearing, Mr. Fulton was in receipt of ODSP. He has lived in both “owner-shared” or s. 21(1) housing, where he would share a kitchen and bathroom with the owner, and housing shared with other tenants that falls outside s. 21(1). Mr. Fulton’s evidence was that outside of public housing, this type of accommodation is the only type that is affordable for him. He is on the waitlist for public housing, and has been advised the wait could be a decade.
[16] Several times when trying to rent a room, when he arrived to view the accommodation with his white cane and a sighted friend, Mr. Fulton was suddenly told the room was no longer available. Sometimes landlords told him they did not want to rent to a blind person. Some landlords expressed concern that Mr. Fulton would not be able to manage living independently in their accommodation. At least one landlord required him to sign a waiver of liability.
[17] Tiring of being told a unit was no longer available when the landlord met him and realised he was blind, Mr. Fulton began disclosing his disability before visiting advertised accommodation. In about October 2017, he and the respondent Hemei Guan exchanged text messages about a room in her home that she had for rent. When Mr. Fulton disclosed to Ms. Guan that he was blind, she advised him that the room was not suitable for him. Ms. Guan said she was concerned Mr. Fulton would have an accident.
[18] Mr. Fulton started an application to the HRTO, alleging Ms. Guan had discriminated against him.
Relevant provisions of the Code and the Charter
[19] The Code’s objectives are significant. As set out in its preamble, the Code is intended to “extend the protection of human rights in Ontario”. It is predicated on the “recognition of the inherent dignity and the equal and inalienable rights of all members of the human family”. The preamble goes on to identify that
[I]t 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 the development and well-being of the community and the Province.
[20] The starting point of the Code, its first provision, states:
Every person has a right to equal treatment with respect to services, goods and facilities, without discrimination because of race, ancestry, place of origin, colour, ethnic origin, citizenship, creed, sex, sexual orientation, gender identity, gender expression, age, marital status, family status or disability.
[21] To achieve its objectives, the Code places limits on the behavior of private individuals. Living accommodation is an area to which the right to equal treatment applies. Subsection 2(1) of the Code provides:
2 (1) Every person has a right to equal treatment with respect to the occupancy of accommodation, without discrimination because of race, ancestry, place of origin, colour, ethnic origin, citizenship, creed, sex, sexual orientation, gender identity, gender expression, age, marital status, family status, disability or the receipt of public assistance.
[22] However, subsection 21(1) of the Code excludes certain accommodation from the right to equal treatment. It provides:
21 (1) The right under section 2 to equal treatment with respect to the occupancy of residential accommodation without discrimination is not infringed by discrimination where the residential accommodation is in a dwelling in which the owner or his or her family reside if the occupant or occupants of the residential accommodation are required to share a bathroom or kitchen facility with the owner or family of the owner.
[23] I will refer to this kind of housing throughout these reasons as “s. 21(1) accommodation” or “s. 21(1) housing”. Much of the evidence in this application refers to “shared accommodation” or “shared housing”, given the available data. Section 21(1) housing is a subset of shared housing. In addition to s. 21(1) housing, shared housing includes such housing as rooming houses, shelters, and hostels.
[24] As noted, Mr. Fulton brings this application under section 15 of the Charter, which provides as follows:
- (1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.
(2) Subsection (1) does not preclude any law, program or activity that has as its object the amelioration of conditions of disadvantaged individuals or groups including those that are disadvantaged because of race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.
Issue one: Was the tribunal correct in determining that the applicant’s evidence did not support a finding or inference that s. 21(1) has a disproportionate impact on individuals with disabilities, and therefore did not satisfy the first step of the s. 15(1) analysis?
Decision under review
[25] Mr. Fulton filed his application to the HRTO on January 29, 2018. In May 2018, the HRTO issued a case assessment direction that identified the housing at issue as falling within the exemption in s. 21(1) of the Code. The HRTO directed that a preliminary hearing be held to determine whether the exemption applied. There was a five-day preliminary hearing in May 2023.
[26] The tribunal correctly noted that it could not make a general declaration of constitutional invalidity. However, the tribunal was invited to decline to apply s. 21(1) to Mr. Fulton’s application on constitutional grounds. The tribunal did not agree to do so. The tribunal found that the applicant had not established a basis for that relief, and issued its decision dismissing the application in May 2025.
[27] The HRTO noted that to succeed on his s. 15(1) argument, the claimant was required to demonstrate two things. First, he had to show that, on its face or in its impact, s. 21(1) creates a distinction based on enumerated or analogous grounds. Second, the claimant must show that the impugned law imposes a burden or denies a benefit in a way that reinforces, perpetuates, or exacerbates disadvantage: R. v. Sharma, 2022 SCC 39, [2022] 3 S.C.R. 147, at para. 28.
[28] The HRTO found that s. 21(1) is facially neutral because it applies equally to all Ontarians regardless of their Code-enumerated grounds. The tribunal reasoned that s. 21(1) makes no distinction between those with and without disabilities and therefore does not directly discriminate against the applicant. Section 21(1) defines the scope of the guarantee from discrimination in housing. The provision excludes a certain kind of dwelling, not a particular class of persons.
[29] Having found the provision to be facially neutral, the HRTO went on to consider whether s. 21(1) has a disproportionate impact on people with disabilities.
[30] The HRTO noted that to establish adverse effect discrimination, the applicant was required to present sufficient evidence to prove that the law creates or contributes to a disproportionate impact on the basis of a protected ground.
[31] Among other evidence put forward at the hearing, both the applicant and Ontario called expert witnesses.
[32] The first of the applicant’s experts was Esther Ignagni, a university professor who was qualified to provide opinion evidence “in the area of disability studies and the barriers to accessing adequate housing faced by disabled people.” Dr. Ignagni testified that disabled people, particularly those who occupy multiple marginalized social positions, are disproportionately affected by housing policies that limit their housing options. However, there are few quantitative analyses examining disabled Canadians’ housing conditions. There is a poverty of data in this regard. The absence of disaggregated data for people with disabilities worsens their vulnerabilities. Dr. Ignagni testified that one cannot generalize in a statistical manner from a qualitative study, but such a study can provide insight and contextual information.
[33] The tribunal then heard from John Stapleton, a senior researcher who was qualified as an expert on statistical analysis, social assistance policy, and poverty issues. Mr. Stapleton’s evidence was that to the best of his knowledge, there is no independent research or academic literature that tracks quantitative data on the prevalence of owner-shared housing in Ontario’s housing market. While boarding houses are not necessarily s. 21(1) housing, in the existing data, they are a close proxy to demonstrate that lower income Ontarians with disabilities tend to disproportionately rent housing that is shared with other people.
[34] Next, the applicant called Adam LaForest, a community support worker and registered social worker. Mr. LaForest was qualified as an expert in the low-income housing market in the catchment area of the Scarborough Community Legal Services and the Greater Toronto Area and the housing needs of low-income individuals in these areas. Mr. LaForest provided evidence about the difficulties faced by his disabled clients in renting housing, and about this population’s need for affordable accommodation, which is often a shared room in owner-occupied housing.
[35] Fourth, the applicant tendered the evidence of Dale Whitmore, the Director of Strategic Litigation for the Canadian Centre for Housing Rights, who was qualified as an expert in housing and human rights issues, including with respect to systemic barriers to housing faced by disadvantaged groups. The tribunal excluded a portion of Mr. Whitmore’s affidavit, which provided contextual evidence about the need for and barriers to affordable housing for disabled people, among others. While the tribunal stated it would provide reasons for excluding this evidence in its final decision, it failed to do so. Mr. Whitmore provided evidence that clients who share a kitchen or bathroom with their landlord are more likely to experience discrimination on the basis of Code-protected grounds, including disability, compared to clients who do not live with their landlord, and that any discrimination in the rental housing market disproportionately affects Ontarians on social assistance and Ontarians with disabilities.
[36] The Attorney-General adduced evidence from two experts. First was Frank Clayton, who was qualified as an expert in urban and real estate economics. Mr. Clayton’s evidence was that housing affordability in Ontario would be increased by a significant increase in housing supply. He opined that Mr. Stapleton’s evidence did not provide meaningful information on occupants of s. 21(1) housing.
[37] The Attorney-General also led evidence from Nathaniel Lauster, a sociologist and housing scholar who provided his opinion about the potential impacts on the rental market if s. 21(1) housing were no longer available.
[38] In considering the evidence before it, the HRTO noted that there was ample evidence of barriers in the housing sector that exist for people with disabilities.
[39] The tribunal accepted that people with disabilities experience socio-economic disadvantage, tend to have lower incomes, and are more likely to have a core housing need, which exists when people do not live in acceptable housing or pay a disproportionate share of their income for inadequate shelter. However, the tribunal found there was not clear and cogent evidence before it to support a finding or an inference that s. 21(1) has a disproportionate impact on individuals with disabilities including the applicant.
[40] In reaching its conclusions, the tribunal did not refer to Dr. Ignagni or her evidence. It found that Mr. LaForest’s opinion did not support the conclusion that people with disabilities disproportionately rely on s. 21(1) housing. With respect to Mr. Whitmore, the tribunal stated simply that it found his evidence should be afforded no weight. It found that his evidence lacked specificity and was anecdotal.
[41] The tribunal found that neither Mr. Clayton nor Mr. Stapleton provided “direct or reasonable proxy evidence on the prevalence of individuals with disabilities living in s. 21(1) housing”. The tribunal noted that Mr. Stapleton’s evidence did not disaggregate s. 21(1) housing from other types of shared accommodation, and therefore did not provide a reasonable proxy for the prevalence of people with disabilities living in s. 21(1) housing as compared to other low-income Ontarians. The tribunal concluded that Mr. Stapelton’s evidence lacked the specificity required to reliably support his conclusions. The tribunal was not satisfied that Mr. Stapleton’s data were reliable.
[42] Similarly, the tribunal found Mr. Clayton’s data were not disaggregated and therefore did not speak to the relative prevalence of people with disabilities in s. 21(1) housing. The data on which Mr. Clayton relied was with respect to boarders, a class of renters that the tribunal found to be both over- and under-inclusive of those who fit within s. 21(1) housing. The tribunal therefore found that the boarder data did not provide a logical proxy for s. 21(1) renters.
[43] The HRTO therefore concluded that the evidence did not demonstrate that s. 21(1), being facially neutral, creates or contributes to a disproportionate impact on individuals with disabilities. Subsection 21(1) therefore applied to Mr. Fulton’s application. On this basis, the HRTO dismissed the application.
The parties’ positions before the court
[44] The parties agree on the general principles that govern a s. 15(1) analysis, but disagree about how whether the tribunal properly applied them. Those general principles include the following:
a. Section 15 guarantees substantive, not formal, equality: Sharma, at para. 37; Fraser v. Canada (Attorney General), 2020 SCC 28, [2020] 3 S.C.R. 113, at para. 42.
b. The claimant must show that, on its face or in its impact, the impugned law creates a distinction based on enumerated or analogous grounds. Second, the claimant must show that the impugned law imposes a burden or denies a benefit in a way that reinforces, perpetuates, or exacerbates disadvantage: Sharma, at para. 28.
c. While the two steps are not watertight compartments, they are to be analysed separately: Sharma, at para. 30.
d. To succeed under the first step[1], the applicant must show that the impact on him is disproportionate to the impact on other groups or the general population: Sharma, at paras. 31, 40. Where a law appears facially neutral, a claimant must provide evidence to prove the impugned law creates or contributes to a disproportionate impact on the basis of a protected ground. The claimant must establish a link between the impugned law and the discriminatory impact. While the impugned provision need not be the only cause, or even the dominant cause, of disproportionate impact, it must be a contributing cause: Sharma, at paras. 42, 44, and 45; Jacob v. Canada (Attorney General), 2024 ONCA 648, at paras. 77-78, leave to appeal refused, [2024] S.C.C.A. No. 488; Quebec (Attorney General) v. Kanyinda, 2026 SCC 7, 511 D.L.R. (4th) 201, at para. 50; Fraser, at paras. 50-52.
e. A disproportionate impact may arise when a restriction that seems neutral operates as “built-in headwinds” for members of protected groups, or when there is an absence of accommodation for a protected group: Fraser, at paras. 53-54; Jacob, at para. 78.
f. In the s. 15(1) analysis, the court must pay attention to the full context of the claimant’s situation and the law’s impact on that situation: Jacob, at paras. 59-60, 68; Fraser, at para. 42.
[45] As noted above, Mr. Fulton had asked the tribunal to decline to apply s. 21(1) to his claim of discrimination. Before this court, Mr. Fulton submits the HRTO ought to have acceded to this request, on a proper s. 15(1) analysis.
[46] Mr. Fulton submits that he demonstrated that people with disabilities, including him, are disproportionately affected by s. 21(1). He submits that, like in Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493, s. 21(1) has a measure of formal equality on its surface - it excludes everyone from the Code's protection. However, not everyone needs to access s. 21(1) accommodations in the same manner at the same rates. He submits that, in accordance with Vriend, s. 21(1) has a substantive discriminatory impact on him and others with disabilities.
[47] The applicant further submits that the tribunal erred when considering the evidence of adverse impact. Specifically, he submits that the tribunal failed to meaningfully engage with the evidence and draw inferences from it, wrongly required statistical evidence when none was available, and wrongly equated a lack of prevalence of people with disabilities in s. 21(1) housing with a lack of adverse impact.
[48] According to the applicant, the evidence he provided is proxy data from which the tribunal ought to have extrapolated. In addition, he provided broader quantitative and qualitative data about housing discrimination faced by people with disabilities. Taken together, he submits that the evidence meets its burden at the first stage of the s. 15 test because people with disabilities are disproportionately: a) low-income earners; b) in core housing need; c) forced to rely on low rent shared accommodation, some of which falls within s. 21(1); and d) subjected to pervasive discrimination in the rental market.
[49] The applicant submits that the HRTO ought to have concluded from the evidence that permitting discrimination under s. 21(1) creates a significant barrier for people with disabilities to access affordable housing and has a disproportionate adverse impact on them.
[50] The Commission supports the applicant’s position. The Commission emphasizes the purpose of the Code is to ensure all members of society are treated with respect and dignity, without discrimination.
[51] The applicant and the Commission submit that the applicant need not demonstrate he is adversely affected as compared to members of other Code-protected groups. Rather, they submit that the disproportionate impact, when caused by an exemption to legislation that is otherwise ameliorative, can be established by comparing the impact to a group that does not suffer from discrimination: Vriend; Quebec (Attorney General) v. Alliance du personnel professionnel et technique de la santé et des services sociaux, 2018 SCC 17, [2018] 1 SCR 464. Whether other protected groups are also excluded from the Code's protections is irrelevant to the analysis; Kanyinda at para. 44.
[52] The Attorney General submits that the tribunal correctly determined that s. 21(1) is a facially neutral provision that does not draw a distinction between those with and without disabilities. The provision simply defines the extent to which the Code protects against discrimination in housing. The position taken by the applicant and the Commission, namely that the definition of discrimination is underinclusive, amounts to an argument that the Legislature should extend the scope of Code protection.
[53] The Attorney General further submits that “there was no evidence before the Tribunal about the disaggregated demographics of persons who live in or seek to live in s. 21(1) housing, nor the effects of s. 21(1) specifically on persons with disabilities.” The Court of appeal in Jacob recently reiterated the need for clear evidence of disparate impact and the insufficiency of a “web of instinct” to pass the s. 15(1) test. The HRTO was correct in finding the applicant did not meet the test in this case.
[54] Ms. Guan submits that the evidence indicates that there is a large supply of shared accommodation that is not s. 21(1) housing that would be available to Mr. Fulton and is priced similarly to s. 21(1) housing.
[55] Therefore, in addition to denying that she engaged in any discrimination toward Mr. Fulton, Ms. Guan denies that Mr. Fulton and other people with disabilities are forced to seek s. 21(1) housing as their only option.
Analysis
[56] I agree with the applicant and the Commission that the HRTO erred in its consideration of the evidence under the first part of the s. 15 test.
[57] In reaching this conclusion, I have not considered the supplementary evidence the applicant sought to file before this court. Those materials consist of information from a City of Toronto website that provides data about average market rents in 2025, and a report published by the Canadian Centre for Housing Rights and Brock University entitled “Research Report Measuring Discrimination in Rental Housing Across Canada”. The Attorney General objects to the materials and submits the court should not accept upon them.
[58] An application for judicial review is generally decided on the basis of the record that was before the initial decision-maker. In my view, the applicant’s proposed supplementary materials meet none of the limited circumstances in which supplementary evidence may be permitted, as set out in Keeprite Workers’ Independent Union v. Keeprite Products Ltd. (1980), 1980 CanLII 1877 (ON CA), 29 O.R. (2d) 513 (C.A.). I therefore have not considered them.
[59] The issues advanced by the applicant and the Commission are as follows:
(i) Whether the tribunal erred by treating a lack of statistics as an insurmountable barrier to satisfying the first part of the s. 15 test;
(ii) Whether the tribunal erred in refusing to draw appropriate inferences from the substantial evidence put forward by the applicant; and
(iii) Whether the tribunal erred in equating a lack of prevalence with a lack of adverse impact.
(i) Whether the tribunal erred by treating a lack of certain statistics as an insurmountable barrier to satisfying the first part of the s. 15 test
[60] The parties agree that there is no precise disaggregated data about the number of Ontario rental units that fall within s. 21(1), or about the number of people with disabilities who live in s. 21(1) housing. It is a situation of “data poverty”.
[61] In reaching its conclusion that the applicant had not demonstrated that s. 21(1) disproportionately impacts individuals with disabilities, the tribunal was critical of this lack of disaggregated statistical evidence. The tribunal noted that while Mr. Stapleton’s data were disaggregated as between individuals receiving Ontario Works and those receiving Ontario Disability Support, the data did not disaggregate s. 21(1) housing from other types of shared accommodation.
[62] The tribunal rejected Mr. Stapleton’s proposed use of boarders as a reasonable proxy for individuals living in s. 21(1) housing. At paragraph 54 of its reasons, the tribunal stated as follows:
Other than Stapleton’s anecdotal evidence that the number is “significant”, I have no evidence on what percentage of boarder housing falls within the scope of s. 21(1) housing. If the number is 90% there may be support for a finding that, at least during the period covered by the MCSS report, persons with disabilities relied more heavily on s. 21(1) housing than did other low-income Ontarians. However, if the number is 25% percent [sic], this conclusion is significantly less reliable. Even if the number is 90%, support for Stapleton’s conclusion could be contingent on the data about the relative reliance of ODSP versus OW recipients on s. 21(1) housing that is not encompassed within the definition of “boarder” in the MCCSS report. I have no evidence on the proportion of this excluded subset relative to all the s. 21(1) housing in Ontario.
[63] With respect to Mr. Clayton, the tribunal accepted that his data accurately captured the prevalence of s. 21(1) housing. However, the tribunal stated:
That said, Clayton’s data was not disaggregated between persons with disabilities and other groups and for this reason, does not assist me in determining whether s. 21(1) of the Code causes or contributes to a disproportionate impact on persons with disabilities.
[64] In a s. 15(1) analysis, no specific form of evidence is required, and the causal connection may be satisfied by reasonable inference. While both statistical evidence of disparity and evidence of broader group disadvantage may demonstrate disparate impact, statistical evidence may not exist, and neither type of evidence is mandatory: Fraser, at paras. 61, 66 and 67; Sharma, at para. 49, Kanyinda, at paras. 52 and 54. The particular evidentiary burden will depend on the claim: Sharma, at para. 49; Kanyinda, at para. 54. To require statistical evidence – in this case disaggregated statistical evidence - where none exists is to place an impossible burden on the applicant.
[65] The tribunal treated the absence of disaggregated statistical evidence as fatal to Mr. Fulton’s application. It found that because there were no specific, disaggregated statistical data available about disabled people in s. 21(1) housing, there was no basis upon which it could draw conclusions or base inferences. The tribunal erred in creating a “rigid evidentiary requirement” (Kanyinda, at para. 60) by insisting on statistical data where all parties agreed none were available, and by allowing the absence of disaggregated statistics to put an end to its inquiry.
(ii) Whether the tribunal erred in refusing to draw appropriate inferences from the substantial evidence put forward by the applicant
[66] The tribunal drew only very limited inferences from the evidence. It held as follows:
73In the record before me, I have ample evidence of the physical, social, cultural or other barriers that persons with disabilities face in the housing sector. I accept that people with disabilities experience pre-existing socio-economic disadvantage, tend to have lower incomes, and are more likely to have a core housing need. However, I do not have clear and cogent evidence that directly establishes or from which I can draw a reasonable inference, that s. 21(1) disproportionately impacts individuals with disabilities.
[67] The tribunal was required to consider the inferences that flow from the evidence the parties adduced. In my view, the tribunal improperly ignored evidence it was required to consider about the “full context of the claimant group’s situation” and about “the outcomes that the impugned law or policy . . . has produced in practice”: Sharma, at para. 49. It ignored “evidence about the physical, social, cultural or other barriers which provide the ‘full context of the claimant group’s situation’”: Kanyinda, at para. 51 [internal citations omitted].
[68] For example, the tribunal wholly ignored Dr. Ignagni’s evidence. She testified that the absence of disaggregated data for people with disabilities worsens their vulnerabilities. The tribunal did not consider that, in the case of acknowledged data poverty, qualitative evidence and available inferences may take on greater importance and require close examination. The tribunal did not refer at all to the evidence of Dr. Ignagni, or Mr. Fulton, and did not consider Mr. Whitmore’s contextual evidence about disabled people’s need for and barriers to affordable housing. Yet Dr. Ignagni provided evidence that although one cannot generalize in a statistical manner from a qualitative study, one can gain insight and contextual information from it. This was important evidence that the tribunal wholly ignored.
[69] The tribunal had evidence before it that established the following: a) shared accommodation, including s. 21(1) housing, is generally the least expensive type of housing available in Ontario; b) people with disabilities are disproportionately over-represented in Ontario’s low income earners; c) people with disabilities therefore have a disproportionate need of less expensive housing options than non-disabled people; and d) people with disabilities are over-represented in housing discrimination complaints. There was also specific evidence from Mr. Fulton and Mr. LaForest about people being denied s. 21(1) accommodation because of their disabilities.
[70] Therefore, a logical inference is that people with disabilities are disproportionately affected by a provision that permits discrimination in a category of housing that includes its most affordable options. Whether other protected groups are also excluded from the Code's protections is irrelevant to the analysis: Kanyinda, at para. 44.
[71] I do not agree with the Attorney-General’s submission that the evidence proffered by the applicant was simply a proxy for poverty, assumptions based on poverty, or pre-existing socio-economic disadvantage. The evidence was not indirect and generic evidence of socioeconomic disadvantage, as was the case in Begum, a case on which the Attorney-General relies. Rather, the tribunal had evidence about the experiences of people with disabilities being reliant upon, and often unable to obtain access to, s. 21(1) housing because of their disability.
[72] I also do not accept the Attorney General’s submission that the evidence was limited to creating a “web of instincts”, which the courts cautioned against in Jacob and Kahkewistahaw First Nation v. Taypotat, 2015 SCC 30, [2015] 2 S.C.R. 548, at para. 34. Instinct is not the same as available inferences. It would be improper to rely on instinct to conclude, in the absence of evidence, that it seems that people with disabilities experience an adverse impact from permitted discrimination in a segment of affordable housing. However, it is not an error to draw inferences from a series of propositions that are grounded in the evidence to find that impact.
[73] I conclude that the tribunal erred in finding the evidence insufficient to ground inferences that lead to the conclusion that s. 21(1) causes or contributes to a distinction through its disproportionate or differential impact on people with disabilities.
(iii) Whether the tribunal erred in equating a lack of prevalence with a lack of adverse impact
[74] I also find that the tribunal erred in equating a lack of prevalence with a lack of adverse impact. I have referred in paragraph 62 above to paragraph 54 of the tribunal’s reasons, where the tribunal expressed its dissatisfaction with Mr. Stapleton’s statistics. In that same paragraph, the tribunal made it clear that in its view, the prevalence of people with disabilities in s. 21(1) housing was an essential, and missing, piece of evidence. This was an error.
[75] As the tribunal notes, there were no reliable available data about prevalence of disabled people in s. 21(1) housing. However, even if there were data that showed a lack of prevalence of this population in s. 21(1) housing, the available qualitative data suggest that this may be precisely because people with disabilities are discriminated against when seeking s. 21(1) housing or other shared housing. As the applicant points out, in British Columbia (Public Service Employee Relations Commission) v. BCGSEU, 1999 CanLII 652 (SCC), [1999] 3 S.C.R. 3, women comprised only 5% of firefighters. However, this low prevalence did not mean the aerobic fitness standard that was being challenged had no adverse impact on women. Presumably, the low prevalence of women from the profession could be seen as a result of that very standard.
Conclusion
[76] The tribunal’s focus on statistical evidence, the lack of disaggregated data, and the lack of evidence of prevalence of disabled people in s. 21(1) housing, when all parties agreed that disaggregated statistical evidence did not exist, created an insurmountable hurdle for the applicant. In failing to consider the inferences available to it from the evidence before it, the tribunal failed to adopt the required flexible approach tailored to the circumstances of this case, and failed to consider whether the evidence demonstrates the type of “built-in headwinds” referred to by the Supreme Court in Fraser.
[77] I pause here to address specifically two of the Attorney General’s submissions. First, the Attorney General submitted that there is no evidence about the effects of s. 21(1) on persons with disabilities. This submission, like the tribunal’s decision, utterly ignores the qualitative evidence that was before the tribunal. It ignores, as did the tribunal, the evidence that establishes that data poverty worsens the vulnerabilities of disabled people. It ignores that, in the context of statistical data poverty, qualitative and contextual evidence bear close consideration. It ignores the requirement to take a flexible, context-specific approach to the evidence.
[78] Second, the Attorney General submits that, because Mr. Fulton had been able to secure shared accommodation, his housing options were not limited to s. 21(1) accommodations. In my view, Mr. Fulton did not have to demonstrate that he would be permanently unhoused if he were not able to secure s. 21(1) accommodation. The difficulties he had securing suitable affordable accommodations were many, and were directly linked to his disability. This is sufficient. The fact that he kept trying to find accommodation and eventually succeeded does not erase the difficulties he and other disabled people face when seeking s. 21(1) accommodation.
[79] For these reasons, I conclude that the HRTO erred and, applying proper principles, the applicant’s evidence meets the first part of the s. 15(1) test, sufficient to demonstrate that s. 21(1) has a disproportionate impact on him.
Issue two: What is the appropriate remedy?
[80] The HRTO, having found the applicant failed at the first step of the s. 15(1) analysis, did not consider the second step or s. 1 of the Charter. The applicant requests that this court, if it overturns the HRTO decision, conduct the remainder of the Charter analysis. He submits there has been substantial delay and the full record is available to the court.
[81] This court is a court of review. The court does not have the benefit of the HRTO analysis on the application of the second step or s. 1 of the Charter to its home statute. While the delay in releasing the decision was unfortunate, the delay alone does not outweigh the benefits of the tribunal making the first determination of the issue.
[82] I would therefore decline to undertake an analysis of the remainder of the s. 15 test and of s. 1.
[83] I would remit the matter to the HRTO to undertake an analysis of the second step of the section 15 test and, if necessary, a s. 1 analysis.
[84] These reasons should not be read as making any comment on the likely outcome of success on the second step of the s. 15 test or any s. 1 Charter analysis that may be undertaken. For example, given the limited scope of these reasons, I have not referred to the landlord’s submissions or some of the evidence that focused on matters that may be relevant to a s. 1 analysis.
[85] In accordance with the parties’ agreement, I would order no costs of the application.
_______________________________
L. Brownstone J.
I agree _______________________________
Matheson J.
I agree _______________________________
Muszynski J.
Released: August 7, 2026
CITATION: Fulton v. Guan et al, 2026 ONSC 2757
DIVISIONAL COURT FILE NO.: DC-25-00000437-00JR
DATE: 20260807
ONTARIO
SUPERIOR COURT OF JUSTICE
DIVISIONAL COURT
Matheson, Muszynski, and Brownstone JJ.
BETWEEN:
Benjamin Fulton
Applicant
– and –
Hemei Guan, Xiaojun Tao, Ontario Human Rights Commission and Human Rights Tribunal of Ontario
Respondents
JUDGMENT
L. BROWNSTONE
Released: August 7, 2026
1Because the HRTO decided the matter on the first step of the analysis, these reasons focus on that step.