Manvi Pinnapereddy is a rising senior at Union High School and a 2025 TURC Junior Scholar. Her research is focused on investigating how different neighborhood demographics impact the level of investment needed to remedy blighted homes.
Mark your calendars for the final presentation of the TURC Housing Policy Research team on August 8th from 11:30am-1pm. You are invited to join us for a lunch and learn event as the students showcase their research findings in Helmerich Hall, Room 219 (2900 E 5th Street). Please RSVP so that we have an accurate count for lunch: https://pp.events/bVQqqkXj
Comparing how factors from high and low-income neighborhoods impact investment opportunities for blighted homes in Tulsa
by: Manvi Pinnapereddy
Blighted homes (also known as vacant, abandoned, and deteriorated homes, VAD) are found in many areas across the United States, leading to pressing issues. The goal of this project is to apply Artificial Intelligence (AI) in financial modeling to explore zoning code revisions, financial feasibility analysis, match housing plans with local incentives and other things that could help redevelop blighted property in the Tulsa area more possible. Specifically, I am working on identifying why and how variables such as median income, average home value, vacancy rate, and population growth affect how much investment a neighborhood needs to remedy housing blight.
I will be using n8n, which is an open-source workflow automation tool, with the goal of creating an investment score for neighborhoods of varying economic status, as well as providing the reasoning for those scores. The high-income neighborhoods I will be analyzing are Brookside, Rolling Oaks, and Midtown. Conversely, the low-income neighborhoods I will be analyzing are Kendall-Whitter, Heller Park, and Collegiate Square. I will be getting an investment score by making a scoring guideline that will be plugged into the workflow, resulting in a score of 0-100. Next, the AI I will be connecting the workflow to is OpenRouter, which will analyze the data I input on median income, average home value, vacancy rate, and population growth to output how these factors affect investment rates, as well as help compare any similarities and differences the neighborhoods have between each other. Hopefully, the results will demonstrate which places, of any income, are best to invest funds to remedy blighted homes in for better community outcomes.
When using n8n to retrieve the investment scores, I found that the low-income neighborhoods had higher potential for investment and the higher income neighborhoods had lower potential for effective investment. I am now going to break down which factors lead each area to have a higher score or what led them to have a lower score.
When asking the AI agent (OpenRouter) why the low-income neighborhoods had higher scores, there was insight specifically on how homes with lower average values have potential to be reestablished and sold for a greater price than the amount that was originally invested. This allows investors to be willing to invest, and consequentially more accessible housing would become available to the public, as renovated houses in low-income neighborhoods are still more affordable than houses in high-income neighborhoods. This does not mean there are no downsides to investing in low-income areas; although not a variable in the original workflow, OpenRouter also lists some risks of higher crime rates in nearby areas and other factors, possibly making the low-income neighborhoods less attractive to buyers. Comparatively, high-income neighborhoods typically have less crime rates per OpenRouter, but it would cost around the same amount of money to redevelop one blighted home in a high-income neighborhood as multiple blighted homes in a low-income neighborhood.
Next, when OpenRouter took vacancy rates into consideration to indicate the opportunities for investment, the neighborhoods with higher vacancy rates (more blighted properties) had more investment possibilities. Oftentimes, low-income neighborhoods have these higher vacancy rates. Although high vacancy rate can correlate with underlying issues like economic decline or a lack of demand for housing, which can lead to more properties becoming blighted, high vacancies may suggest that properties are underutilized, offering the chance to revitalize and improve the living conditions of the area. Vacant properties often have lower acquisition and renovation costs, allowing investors to allocate funds to more impactful community projects related to infrastructure and social services. Investing in these areas not only addresses housing shortages but also fosters economic growth and helps build stronger, more cohesive communities in low-income areas by redistributing populations to these areas (noted by AI system by Cypher Labs). On the other hand, while low vacancy rate areas are in higher demand, it is far more expensive to remedy blighted homes in these areas than high vacancy neighborhoods, leading to a higher investment score for low-income neighborhoods.
Accordingly, when population growth is taken into consideration, higher population growth occurs in low-income neighborhoods, indicating areas with strong potential for impactful investment, lower costs early on, and the ability to create meaningful positive changes for many residents. Meanwhile, low population growth in a high-income neighborhoods signal potential for redevelopment, but investors must weigh these opportunities with the need for affordability and community engagement to ensure successful and inclusive reinvestment (AI system by Cypher Labs). Future directions for these assessments and workflows include adding more variables to consider articulating a more multifaceted answer on why various areas lead to higher or lower investment. I would also like to add more neighborhoods, perhaps some middle-income ones as well as calculate the actual average amount of money it would take to renovate or redevelop one home in a high-income area versus one home in a low-income area.