Over the course of the semester, I took an interest in affordable housing in NJ and the rich datasets that accompany the now 46-year history of the Mount Laurel Doctrine. Because the project adopted a story element that includes a crash course on the Mount Laurel Doctrine, I won’t summarize that history here. Early on I imagined a dashboard visualization for this project that would allow people to jump right into the data. But after researching this topic throughout the semester, I came to the conclusion that a story-telling approach would work better for two primary reasons.
First, I quickly observed that the data was intrinsically linked to a complex web of four primary court rulings, the now-defunct Council on Affordable Housing (COAH), the Fair Housing Act, and the Fair Share Housing Center (FSHC) public advocacy organization. Each of these bodies and rulings played a role in developing the various methodologies that ultimately determine how many affordable housing units each NJ municipality must build under Mount Laurel. Their impact on the data is two-dimensional through the development of factors that determine housing calculations, as well as four primary periods of time since 1975 that courts and municipalities use to determine their total affordable housing obligation. While this history could be summarized or generally left out in a data visualization, I worried that surface-level explorations would lead viewers to lack context and meaning behind what they were exploring.
The second rationale for a story approach was the belief that perhaps this small project could help others understand and engage with the Mount Laurel Doctrine in a way I had not seen in my research. I strived to finely balance engagement through visuals, chunking, and interactive visualizations with the dryer complexities and history of the topic.
The process used to create the final story was generally broken into three parts – dataset acquisition and expert interviews, software exploration and application, and academic research into the Mount Laurel Doctrine.
Data Acquesition and Expert Interviews
For datasets and expert interviews, the keystone connection to this project was David Kinsey. Dr. Kinsey is a visiting lecturer at Princeton University, a partner at Kinsey & Hand, Planning, and a veteran public advocate “vindicating constitutional housing obligations and achieving smart growth in NJ.” Dr. Kinsey was kind enough to provide me with the core dataset used in the latest Mount Laurel ruling, a rich 50 tab+ spreadsheet that provides complete transparency into how numerous methodologies are applied to arrive at final housing numbers for each municipality. Dr. Kinsey set me various materials in addition to the core dataset, and provided feedback at a few touchpoints across the semester.
I also had the good fortune to connect with two staff members at the FSHC, Yvette Chen (Planning and Policy Analyst) and Elorm Ocansey (Housing Access Organizer), for interviews. Yvette’s insight into the data led me to narrow my focus on a few factors (developable land, income measures) and helped me understand key concepts (like Cost Burdened status, or when you pay 30%+ of your income towards rent) and how to pull micro-level datasets from the American Community Survey. Elorm widened my perspective beyond the data, telling me how the FSHC helps low to moderate-income people in NJ through the Housing Resource Center and constant pushing against municipalities looking to avoid their fair share responsibilities. Elorm’s feedback led me to become more interested in the advocacy and mission of the FSHC and Mount Laurel, which was ultimately reflected in the story-driven approach of the final project.
Software Exploration and Application
Though a smaller part of the project directly and more a component of the course, learning and applying the software was an important step. One of the largest conclusions I reached was the decision to embed a Carto map within Tableau, which applied each software’s greatest strengths. Carto was able to handle the dense municipality level map data Tableau couldn’t, while Tableau provided the comprehensive suite of functionalities (PowerPoint-like slide creation, creating interactive bar graphs) that enabled me to take a story-driven approach.
This project ended up including considerably more research than I expected, but this part turned out to be my favorite component. Learning about the Mount Laurel Doctrine from the FSHC and academic articles was interesting, but the real depth came from reading the court cases directly. Especially the 2017 Jacobson ruling, which was the first court source document I’ve read. Though dense, the recounting of the expert witnesses (of which Dr. Kinsey was among) was captivating as you follow the two sides of the case and how their proposed methodologies both seem reasonable, yet advocate for opposing views (the municipalities on one side wanting lower numbers, and developers and public advocates (a strange partnership!) seeking fairer numbers that represented the ideals of the original Mount Laurel Doctrine).
This research was ultimately done to both try to grasp some of the complexity of the Mount Laurel Doctrine so that I could summarize it effectively, and find quotes from source material, which I believe lends credibility and weight to the final narrative. Additional research was done primarily in the New York Times physical paper archives to find interesting images and headlines that captured each time period discussed.
In terms of data visualization, the conclusions of this project were less about novel findings and more about peeling back the invisible and complex factors that determine affordable housing obligations for the state. I believe I was only partially successful in this task, enabling users to explore
- Percentages of Low to Moderate Income Households by NJ Region
- Average Income Measures per municipality
- Undeveloped Land (weighted) per municipality
- Final housing numbers across all court-derrived time periods.
What I was hoping to enable was the ability to consider, on a municipality level, what factors were primary and what were secondary in deriving housing numbers. The overlay approach taken on the map somewhat enables this, but you can only really consider 1 factor at a time and its impact on the housing numbers – layering multiple factors brings about color (hue) mixing that would require people to understand, for example, that a dark green municipality suggests that both the orange factor and the blue factor had similar relevancy, and mixed to create green.
Other limitations were simply due to complexity. For example, in the above map, note that Newark (the gray municipality to the east) isn’t pulling any color. This is because Newark is a Qualifying Urban Aid municipality, and is exempted from certain requirements (leading to no yellow or blue coloration because there was no data processed). While I attempt to bring this up in the Legend notes, this is one of a few factors I was not able to include in this map (due to time and Carto layer limitations), which ultimately leads to an incomplete picture of all of the factors that go into the methodology.
Despite these limitations, I do think there was greater success in simplifying the data to provide something understandable. Especially the final slide, which provides a final percent (30%) of progress the state overall has made towards achieving the ideal Mount Laurel fair share obligations through 2025.
Overall I am proud of this project, and plan to send it to the FSHC to see if they might want to use it for something. Perhaps it would make a good blog post for them, assuming my writing and visualizations are accurate. I had two main areas in which I believe this could be improved, in addition to the aforementioned lack of a complete picture of all of the factors that go into affordable housing numbers.
The first area is trying to normalize the data. This was discussed very early on (normalizing housing numbers with municipality populations, so your perceptions of each municipalities fair share is not skewed by high-population areas), but I was not able to perform the calculations in time. Dr. Kinsey’s data only includes Low to moderate-income household calculations at the regional level, so I would have to get the raw American Community Survey data and process it down to find the number of low to moderate-income households per municipality. This could surely be done with more time.
The second area that I would have liked to expand on is providing some narrative to the opposing side (municipalities wanting to build less affordable housing). While many of the rationales are exclusionary and “Not in my backyard,” one critical aspect warrants more discussion. That aspect is that building affordable housing can only be done through developers by law (a municipality cannot build their own housing directly). Because of this, towns must work with developers, who will only build at a 4:1 ratio. That is, for every 4 luxury/normal housing units, 1 affordable housing unit gets built. Developers force this ratio to ensure they still make a profit while building affordable housing. Because of this system, if a town needs to build 100 affordable housing units, it’s likely they actually will need to build closer to 500 total new units. This 5-fold increase in actual housing likely has a much larger impact on communities (sewer, school tax base, infrastructure, density, etc), and is less talked-about in the affordable housing discussion, in my opinion.
Overall I think this project provides a mid-level dive into the Mount Laurel Doctrine through data, history, and visuals, and will hopefully be an interesting read.