Whether and how landlords can use information from an existing or prospective tenant's criminal history was the subject of an April 4, 2016 "guidance" memo from the Department of Housing and Urban Development (HUD). A guidance memo does not have the legal authority of a court decision or a statute, but it carries considerable weight with judges nonetheless. The memo did not come as a surprise, given that it echoes a sister agency's long-standing advice to employers with respect to their use of arrest and conviction records when making employment decisions. (See the Equal Employment and Opportunity Commission's "Enforcement Guidance" for employers.)
Rental applicants and tenants with criminal records have not, in the past, received much protection against housing discrimination. As the HUD memo noted, while federal fair housing laws protect people from discrimination on the basis of race, color, religion, sex, familial status, national origin, and disability, they do not extend protected class status to those with a criminal record.
This has meant that under federal law, landlords were free to adopt screening policies that targeted people with criminal histories (for example, categorically rejecting rental applicants who had an arrest and/or a conviction). (Notably, some local laws are more restrictive; for example, San Francisco passed a "Ban the Box" ordinance in 2014 that, among other things, forbids a question about arrests on a housing application for housing subsidized by the city.) The only exception to this practice concerned past drug users; landlords could not reject when past drug users had a conviction for drug use, on the theory that drug use is a disability (which is a protected class). People with convictions for drug sale or manufacture, however, did not enjoy protected status.
Fair housing advocates have for a long time sought to invalidate landlord policies that automatically reject applicants with conviction or even arrest records. But given the lack of "protected class" status, how to do it?
The opening came in a U.S. Supreme Court case in 2015, which held that housing discrimination could be shown not only in cases where the landlord clearly intended to treat one group unjustly or prejudicially, but also by showing that the landlord's policy (although neutral on its face) had the effect of harming people in a protected class. (Texas Dept. of Housing & Community Affairs v. Inclusive Communities Project, Inc., 135 S.Ct. 2507 (2015).)
An example of intentional discrimination by a landlord would be a stated policy of "I do not rent to Hispanics." Indirect discrimination, also known as "disparate impact" discrimination, might be, "I do not rent to people who live in [an area of town densely populated by Hispanics]." In the first instance, the rejected applicant could prove discrimination simply by introducing evidence of the landlord's stated policy. In the second example, the applicant would have to prove that, because an overwhelming number of Hispanics lived in that area of town, the effect (or impact) of the policy was to discriminate against them.
The HUD memo seized on this new, indirect way of showing discrimination, by reciting statistics showing that many, many more Blacks and Hispanics are arrested and convicted than are Whites. So, when a landlord applies a "no arrest or conviction" rule when screening tenants, members of these protected classes are being harmed more frequently than, say, Whites. In other words, such a policy, while neutral on its face, in fact disadvantages, or discriminates against, protected individuals.
The HUD memo did not go so far as to advise landlords against turning away applicants with relevant convictions (more on that below). It did, however, disapprove the practice of rejecting based only on an arrest-without-conviction basis: This is never appropriate, said the memo, because an arrest "shows nothing more than that somebody probably suspected the person apprehended of an offense." (Memo, page 5.) In other words, the arrest is not relevant to the only justifiable reason a landlord might have for such a policy: The desire to keep other residents, guests, repairpersons, and employees safe by rejecting known troublemakers.
As to convictions, however, landlords have more leeway—and quite a challenge ahead of them. Some convictions might indeed be relevant to the landlord's duty to keep residents and others safe. For example, a recent conviction for assaulting a neighbor might justifiably cause a prospective landlord to hesitate, as might a conviction for perpetrating domestic violence at the applicant's last rental. (Note, however, that many states prohibit landlords from denying housing to victims of domestic violence, when that is the only reason for the denial.)
On the other hand, a decades-old conviction, especially for a non-violent crime, probably has little to do with the chances that the applicant will be a problem tenant now.
The HUD memo suggested that landlords consider several factors, such as the nature and age of any conviction, before using it as a basis to deny housing. For example, how old is the conviction, what was it for, what were the circumstances, has the applicant been rehabilitated and crime-free since then, and so on?
As you can see, this inquiry involves the landlord in a case-by-case analysis—one that many landlords may be loath to undertake, because there are no "bright lines" to help them. On the contrary, landlords will have to make judgment calls as to, for instance, how old is old enough—and hope that, if challenged, they stand up in court.