Tuesday, November 19, 2013

Affordability Index: Who is this for, exactly?

The new HUD housing/transport location tool has made a minor ripple in the urbanist blogosphere. The hope, as stated by HUD is to,"...to provide the public with reliable, user-friendly data and resources on combined housing and transportation costs to help consumers, policymakers, and developers make more informed decisions about where to live, work, and invest."  There have already been a couple of blog posts at sites like Next City and DC Streetsblog that have played with the tool and speak on its functionality. I don't wish to reproduce this work, but to ask a more fundamental question: Who is this exactly for and what are we trying to show here?

First, the tool is an excellent visualization application. The display is clean, the controls fairly intuitive and you can get a lot of information in a simple format. Kudos to HUD and its partners for making this available. But I also feel that this tool is supposed to provide a service that does not accurately reflect how people make housing decisions. I get it, this is a visualization application. It's not a policy or an office, but the stated goal of the tool does, in fact, hold a particular political orientation.

Succinctly, the tool, in holding up housing options for people as largely a rational exercise in cost tradeoffs, does nothing to interrogate the ways in which we construct housing or provide transit infrastructure. In framing housing choices as basically a rational choice exercise without also recognizing that most households, including all low to moderate income households, actually have an incredibly restricted set of choices based on market power, the tool subtly shifts the onus of housing affordability onto potential residents. This is not to say that this tool is part of a greater insidious project to disenfranchise marginalized populations, but it does show that the framing around this tool's use is based on a rational consumer in a depoliticized landscape of housing and transportation costs that are simply presented.

More specifically, while the tool offers a set of idealized populations you can map with that can offer some rather stark comparisons (the first image is a map of Portland's affordability index for a "region typical" household making around 56k a year):

It's clear your "region typical" household basically has their pick of spots. This may be the closest one can get to an idealized decision structure for a household. There are clear tradeoffs to be seen, at least for the rental market. When taking home ownership into account the decisions change drastically (again for "region typical" household):

We see a large jump in prices. This is largely a reflection to the really tight housing market we have in the region right now and you can identify some of the areas in the city and outer parts of the region that are still affordable and, in the case of Portland proper, is quickly being converted bought up by better off households.

Now, what does this same market look like for the "low income" category?

The difference is there in stark relief. Low income renters, on average, are looking to spend 54% of household income on housing and transit. More recent rental market changes are especially evident in this map. Northeast and central east Portland are now areas where most low income households can be expected to pay nearly half of their income on housing and transport even though these areas are central to the region. For low income homeowners or prospective homeowners the difference is even more stark:

Uhhhhh...so....yeah. Nothing much to say here, really. If a low income household is able to buy a home (an unlikely proposition given still-tight lending standards and the hot market in PDX) this household can expect to 80% of its income on housing and transit. 

It's clear that low income households, both renters and homeowners, have a much more restrictive set of "options" available to them in terms of housing and transportation. This is to be expected in a system where housing provision is primarily left to market forces. There's nothing really new here, but it's always good to be able to visualize such stark disparity in simple ways. But if this affordability index tool is designed for policymakers and planners, then I'm wondering what this actually adds to planning or policy practice? Planners already know that housing and transportation are large costs and that these costs are borne more heavily on low income households. I don't know if this tool is telling us anything new or unique. If this tool is showing planners something new, then those planners don't deserve the title.

The obvious rejoinder to this observation is that this simply a tool and planners and policymakers can use it different ways. As a visualization tool it can display disparity in stark, unmistakeable terms that can help when engaging with elected officials or the public. Its simple interface allows for folks outside public planning agencies and government to make similar arguments as long as they have access to a decent computer and internet access (a barrier, but one that is being slowly broken down). In this way, it can potentially be useful as a tool for a particular political project, but as an actual analysis tool it is lacking. 

One gap, an entirely understandable one given model constraints but still big, is that you cannot disaggregate households on different demographic characteristics (see methodology here). For example, you cannot separate households by race and compare. Why is this important? Because household income varies widely among different racial/ethnic groups. In Portland, for example, median household income by race (ACS 5yr-2011) taken from Social Explorer:

The regional typical household for Portland makes around 56k a year for the HUD tool. But we can see that the only racial group that comes close to being a "typical" household is that for White households. Admittedly, this is a very White region compared to others, but this difference here is still striking. Planners and policymakers can see that if they're planning for households other than White ones, then they need to really focus on making housing available more generally to low-income populations, in particular Black households. But the tool, as offered, can't show planners that kind of detail or difference. This severely limits its potential as an actual tool of analysis for planners. As offered, the tool does not really give an accurate representation of housing or transport costs except at the most general of levels. 

For equity oriented planners, or for planners who want to be able to better track, analyze and display difference within their cities, this tool obscures much more than it can potentially display. I recognize the limits of creating a standalone web mapping tool, let alone one that is as easy and kind of fun to use as this one, but I have a hard time seeing the upside here. At least when it comes to planners making use of this. Add to that the implicit rational choice orientation of the tool that depoliticizes and obscures the role the market and market actors play in restricting housing to those who can afford it and the tool loses even more value. 

Overall, I give this tool a C- in terms of utility for planners or policymakers looking for analysis. It can maybe help planners to ask some deeper follow up questions on distribution, but, in my opinion, they should already be asking those questions. I'm inclined to give it a B or B+ for visualization in a simple, easy to use web format. What say you all? Is this a move forward? Does it potentially move HUD's mission forward? Does it help planning or planners?