Political Alignment and Housing Transactions
with Yongqiang Chu and Cong (Roman) Wang
Abstract: By matching voter registration records with housing transaction data, we find that voters who share the same political affiliation as the current U.S. president are 0.214 percentage points (10% of the unconditional probability) more likely to engage in housing transactions. Both buying and selling transactions increase, suggesting that reduced perceived economic uncertainty drives this effect. We develop a spatial equilibrium model predicting that these politically motivated transactions lead to lower house prices in aligned neighborhoods due to increased supply without corresponding demand increases. Empirically, house prices in aligned census blocks are 2.0% lower than in misaligned blocks, consistent with our theoretical predictions.
Presentations: UNC Charlotte (2024, 2025), FMA Semi-Finalist Best Paper (2025)
Press: Featured in The Wall Street Journal article, "What Puts Republicans in the Mood to Shop."
When the Train Never Comes: Property Value Impacts from the Announcement and Cancellation of a Light Rail Project.
The Professional Geographer (2025)
with Isabelle Nilsson and Elizabeth C. Delmelle
Abstract: Academics and practitioners have long been interested in the potential property value increases associated with new transit investments. The placement of transit stations is not random, however, and often they are placed in locations most likely to experience growth. Therefore, isolating the impacts of the new infrastructure from these confounding factors has been a challenge. In this study, we examine the effect of the announcement and subsequent cancellation of a transit project on sales prices, taking advantage of a unique natural experiment, the Durham Orange Light Rail Transit project. Using a set of closely matched treatment and control properties, we find no evidence of an announcement or cancellation effect on station-adjacent property sales prices. We find evidence, however, that prices in denser areas, closer to the city center tended to appreciate at a higher rate. These findings suggest a shift in demand for living in dense, urban environments, which is often where rail transit stations are located. The results are essential for urban planning as they show that rail transit is planned in growing cities with tight housing markets where properties, including those located in neighborhoods with planned stations, are already on an upward growth trajectory in terms of price.
Presentations: UNC Charlotte (2023)
The Mere Exposure Effect in Architecture
Humanizing Digital Reality (pp 589–602, 2018)
with Christopher Beorkrem and Jefferson Ellinger
Abstract: Per Schelling’s model of Segregation, the population will innately segregate itself based on preferences, often leading to organization by race and class. This subdividing of communities through segregation increases social tension, discourages communication, and isolates those who are different. In 1968, Robert Zajonc proved that subjects rated a familiar stimulus more positively than similar yet unfamiliar stimuli. The mere exposure effect is a phenomenon by which people develop a preference for things solely because they are familiar with them. Architecture can diminish the impact of social segregation through mere exposure by examining the effects of architectural interventions and programs. Through mere exposure designers can create new connections between members of society by rethinking circulation paths, carefully considering the geolocation of program, and creating more effective public space. By incorporating modern social behavioral analytics into design logics, social spaces can facilitate more productive engagements between occupants. Examining the effect of unit location on circulation and noting the most effective locations for public goods, developers and city planners will increase communication between community members. Increasing communication as a primary goal of design will facilitate the development of stronger communities. Although the tool specifically targets residential complexes, the concept is scalable. Providing a designer an automated method for evaluation and data collection based on the mere exposure effect in urban design and architecture can create more informed design of public space. The goal is to create a more diverse and sustainable community through an informed understanding of how space and program influence behavior.
Presentations: Design Modelling Symposium 2017, UNC Charlotte (2016, 2017)
Partisan (Mis)Alignment and Earnings Forecast: Evidence from Sell-Side Analysts
with Yongqiang Chu and Cong (Roman) Wang
Abstract: How do partisan beliefs impact how analysts price securities? We use voter registration data of analysts to show that analysts whose party is not affiliated with the president have significantly lower earnings forecasts than other analysts. This effect is neutralized when competition increases in the sector. We examine heightened competition geographically by controlling for battleground states during election cycles. Additionally, we test national effects by controlling for greater partisan disagreement as measured in news feeds. We provide evidence that the impact of partisan bias is lessened during periods of heightened competition. Our results are robust to various fixed effects and clustering. They suggest that partisan beliefs manipulate asset prices..
Presentations: UNC Charlotte (2023, 2024)