Thursday, July 24, 2014

Fair housing and the data paradox

by Patrick Reed, National Housing Conference

In a push for crowd-sourced innovation and better transparency, federal agencies are opening up their data
vaults to the public, substantially increasing the accessibility of this information. State agencies and local governments are increasingly following suit. In general, most would agree that improved data accessibility is a good thing; however, increased availability is not without its challenges, particularly for low-income households and their advocates. Consider the propositions of the following paradox:

A. Open access to data improves opportunity for low-income households.
Example: Using publically available data, PHAs can create opportunity maps to locate census tracts with strong social capital and strong transit access. Housing Choice Voucher counselors can use these maps to help inform their clients’ decisions.

B. Open access to data harms opportunity for low-income households.
Example: Using an online tool, a high-income family realizes that a home they are interested in purchasing falls within the boundaries of a census tract with a large number of foreclosures and vacancies. They decide not to pursue the property any further. Current residents in the neighborhood miss out on the addition of a stable family that could have potentially drawn in more resources and higher-income neighbors.

If we’re thinking in terms of net gains and losses, we can’t say that both of these propositions are simultaneously true.

Lisa Prevost's article in last Friday’s New York Times does an excellent job of contextualizing the data access problem. Third-party companies help people access detailed information about neighborhood demographics, income levels, crime, school quality etc. These companies often source their information from publically available database sites like the Census Bureau’s American Fact Finder. As many of these companies aren’t licensed brokers, they aren’t subject to the Fair Housing Act’s laws about steering. Because their products affect the market and competition, licensed brokers are also beginning to test their own boundaries. This raises a number of policy, and ethical, questions, such as:
  • Should non-brokerage companies who capitalize on public data be subject to the same Fair Housing rules?
  • For brokers who don’t want to lose ground to unregulated third-party companies, if a broker directs a client to sites like American Fact Finder (or even links these sites to her firm’s web page), is that a violation of the Fair Housing Act?
  • Should agencies like HUD and the Census Bureau limit data that are sensitive and potentially damaging to the pursuit of integration?
  • Do low-income populations have the same access to information as high-income populations? If so, do they have the resources that will help them use it to their advantage?
I don’t know the answer to these questions, but I think housing advocates and civil rights groups need to substantially raise the profile of this issue before our current “no-questions asked access” becomes the norm—if it isn’t already. Additionally, we need researchers to investigate whether or not the availability of data has affected patterns of racial and economic segregation. Who knows, maybe research findings will help alleviate some concerns.

If you’re still not convinced that this is an issue, perhaps a quick anecdote will help: last week, a peer and I completed a graduate school project that analyzed dollar per square foot costs, walkability and school quality indicators in an increasingly stratifying urban county. Using publically accessible data, GIS and some statistical sorcery, we were able to pinpoint particular properties—not block groups or tracts—in high-quality school catchments where residents weren’t paying the same premiums as their neighbors for a high-quality education. Our purpose was to promote child well-being by creating a “burden-bargain index” that would help CDCs locate target properties for potential acquisition or restoration.

A couple of days after our presentation, I was approached by an individual who works in a mid-management position, is salaried, and is assumedly quite stable: “I heard about your project—sounds neat. My husband and I are in the hunt for a home right now in the county. You’ll have to show me your maps at some point.” 

Given the project’s intent, how should one respond to such requests? How will our policymakers respond?

1 comment:

Anonymous said...

True -- the data overlays are a form of redlining. I encouraged my sister to purchase a newbuilt infill in a neighborhood with high foreclosures --- a private effort to help stabilize.....
Now when I google our address on zillow the first overlay is "crime" (higher in a high unemployment area) .. then school accomplihsments....
personal experience with a broker contacted to assist in renting the upstairs appartment "why would anyone want to pay X (.90/sq/ft) to live in this neighborhood. They could go to ...... lots of showing the apartment but no tenants....
we self-rent via rent.com and are successfull across ethnicities...