Inman

Steering 2.0? Data may undermine fair housing laws

Big data image via Shutterstock.

This is the final installment of a four-part series that explores how the spread of neighborhood data could undermine the role of real estate agents and aggravate social and economic disparities between communities.  

Only every so often will Phil Faranda’s clients say “too much coffee, not enough cream” when he recommends listings in certain neighborhoods. But plenty of buyers in Westchester Country, New York, still shun certain communities due to their ethnic makeups, he said.

That’s made the Realtor anxious about a trend in online real estate: surfacing hyperlocal demographic information.

“I think the public has always been tone deaf in interpreting data and when … you throw into the mix the kind of emotional and passionate attitudes about race and ethnicity … you’ve got a vitriolic mix,” said Faranda, who owns J. Philip Real Estate, a brokerage based in Briarcliff Manor, New York.

Many Americans are inclined to concentrate themselves in socially uniform communities. But a growing number of websites and apps are making it easier than ever by showing data including the percentage of homes with children, ethnic makeup and political leanings of localities.

This level of transparency may undermine the spirit of fair housing laws by fueling people’s tendency to sort themselves along demographic lines — a consequence that would compound the influence of data like online school ratings, granular crime rates and sex offender data on neighborhoods.

This screen shot taken in May 2013 shows two listing results on realtor.com supplemented with block-level demographic data by the browser plug-in CensusConnect. The plug-in no longer works because its creator found it too time-consuming to continually update. 

Some agents, like Faranda and Schaumburg, Illinois- based Lyn Sims, think that it’s inappropriate for sites and apps to surface data they don’t think they’re legally permitted to disclose.

“I think that if I cannot provide neighborhood ethnic makeup, neither should an app or a website,” Sims said.

Others argue that consumers have a right to access such data, regardless of whether they discover it in the middle of filtering for homes online.

“These sort of statistics and information matter to homebuyers and just because I legally am prohibited from providing it to them myself doesn’t make it any less important,” said Alyssa Hellman, an agent in Washington, D.C.

Next-gen search: personalized, contextual and predictive

New data technologies are rolling out at warp speed, challenging laws and best practices and providing consumers a new level of information and granularity.

StreetAdvisor has led the charge in personalizing neighborhood search. Users may filter for neighborhoods based on the density of different categories of people (like “hipsters” and “gay and lesbian”), personality (like “neighborly spirit” and “clean and green”) and “things to do” (like “eating out” and “gym & fitness).

In the near feature, NeighborhoodScout will roll out a revamped search tool sophisticated enough to allow people to “find mobile home parks dominated by Russian speakers,” according to CEO Andrew Schiller.

Screen shots of Dwellr neighborhood profile for a census tract in Manhattan’s Lower East Side.

The feature will complement its patented “Match” tool, which asks users to “specify a neighborhood you love,” and a geographic area of interest, and then returns census tracts that most closely resemble the specified neighborhood.

Dwellr, a mobile app recently introduced by the Census Bureau, queries people on their marital status, education level and “ideal view” among other things, and then matches them with 25 cities based on their answers.

Relocality takes what some call “predictive search” to the next level. The site recommends neighborhoods based on users’ Facebook profiles, going so far as to scan their wall content and photos to determine their preferences.

Such services benefit consumers, and much of the data they display doesn’t push the envelope. But these services, along with other websites and apps, can also help frame the real estate search around demographic characteristics.

Off-limits for agents, fair game for websites?

NeighborhoodScout, Dwellr, PolicyMap and online tools offered by the Census Bureau and the Department of Housing and Urban Development (HUD) are among services that show demographic data at a neighborhood level, including ethnic composition.

And as the browser plug-in CensusConnect demonstrated, sites and apps may now tap the new Census Bureau application programming interfaces (APIs) to show this information all the way down to the block level.

Displaying that data can do “what brokers have been forbidden to do for the better part of four decades, which is steer people and influence their decisions,” Faranda said.

The Fair Housing Act, a federal anti-discrimination law established in 1968, prohibits discrimination in housing-related transactions on the basis of race, color, national origin, religion, sex, disability and familial status.

Many agents shy away from discussing neighborhood demographics, particularly those that even graze the topic of race, out of fear of violating these rules.

What do you do if clients ask you for information about groups protected under fair housing laws? Please tell us in our comments section.

Screen shots of Sitegeist neighborhood profile for a census tract in Manhattan’s Lower East Side. 

According to Laurie Janik, former general counsel for the National Association of Realtors (NAR), the rules all but prohibit agents from ever raising the issue of some neighborhood demographics. And they are “skating on thin ice” if they disclose the information even in response to a question from a client.

“It’s risky,” she said. “Who wants to be named in a lawsuit?”

Asked why Redfin doesn’t display crime or demographic data, a spokesperson said, “As a company of agents, we carefully adhere to fair housing laws involving things like discrimination and steering.”

The sensitive nature of granular ethnic data is partly what prompted the Sunlight Foundation, the creator of neighborhood information app Sitegeist, to remove the data from Sitegeist’s mix of neighborhood information at the last minute, said Jeremy Carbaugh, lead developer for Sitegeist.

Carbaugh said his team has assembled a “long list” of additional information they may integrate into the app, including crime data and languages spoken.

The app already shows the median age and number of children in a census tract, along with political party contributions at the ZIP code level. (Zillow CEO Spencer Rascoff has said that he would like Zillow to show political contributions by household.)

Litigation ahead?

As time passes, services resembling Sitegeist, Dwellr and CensusConnect will stitch hyperlocal demographic information into neighborhood or listing-search experiences, according to Russ Cofano, former general counsel at John L. Scott Real Estate and the current CEO of the Missouri Association of Realtors.

These apps and sites will inevitably spark legal action, forcing courts to determine if tying demographic data to neighborhood or online listing search violates fair housing rules, he said.

“There will be people who push the envelope,” he said, and “fair housing laws will at some point in time have to address how access to data could be used in a way that creates a discriminatory effect.”

It’s easy to imagine that racial data could be the focus of such litigation. Brokerages and listing portals have so far shied away from publishing racial data. (One exception is the brokerage Movoto, which shows racial data at ZIP code level on its neighborhood pages.)

Screen shot showing Movoto’s “race and gender” breakdown of the ZIP code 20020 in Washington, D.C. 

But real estate sites and apps are shining light on data relating to other protected classes.

Zillow offers neighborhood-level charts on relationship status (single, widowed, divorced or married) and the proportion of “homes with kids,” for example. The San Francisco Association of Realtors shows the percentage of family households for all city neighborhoods.

Sites and services that display demographic data could compound the divisive influence of some other data on neighborhoods.

Demographic data can often be an indicator or proxy for neighborhood wealth, meaning people who sort themselves into communities based on neighborhoods’ demographic makeups tend to be sorting themselves along economic lines.

Sex offender, school and crime data, and other forms of neighborhood information have already begun to further pry apart the social fault lines between neighborhoods, pushing up prices for neighborhoods portrayed positively by the data, many of them already affluent, and driving down prices for neighborhoods tarnished by it, many of them already distressed.

If websites and apps can disclose demographic data, do you think agents should be able to as well?

This is the final installment of a four-part series on neighborhood data. Read part 1 on crime data here, part 2 on sex offender data here and part 3 on school data here