Identifying the Most Suitable Counties in America for Residential Ride-Sharing and Real Estate Growth Potential: A Quantitative, Geographical, and Financial Approach

Abstract

Serving as a guide for urban planners, civil engineers, land developers, and corporate executives as they update, plan, and invest in communities throughout our country in light of the impending introduction of autonomous vehicles in the coming years, this thesis conducts a geographical analysis county-by-county across the United States to quantify and study the relationship between ride-sharing potential for individual transportation trips originating or terminating at a personal residence and residential housing prices in local real estate markets. As intelligent automotive systems continue to progress in their development and become incorporated into our society, individuals will eventually be presented with opportunities to transport themselves using autonomous vehicles and to share rides to enhance their mobility, thereby leaving effects on not only the future of transportation but on many sectors of our economy, from business to real estate and from energy consumption to more efficient use of transportation infrastructure. Besides making transportation more efficient, these smart-driving systems will increase accessibility to all areas of the country, offering opportunities for individuals to mobilize themselves from their homes who were previously unable to do so, allowing individuals to make more productive use of their commuting time from their places of residence to final destinations, and making transportation of goods, services, consumers, and producers more efficient. This increase in accessibility and efficiency will influence the decisions of producers who are looking for the best place to conduct business and of individuals about where they would like to establish their residences, thereby affecting the desirability for working and living in particular areas of the country. In this thesis, we focus on the relationship between residential ride-sharing potential in each county of the United States and its corresponding real estate market. Inspired by previous ride-sharing studies that focused on particular cities or geographical areas, this thesis orchestrates ride-sharing analysis on a particular type of trip, namely trips originating or terminating at individual residences, and does so across every county in America. Using ride-sharing algorithms, Monte Carlo simulations, and data analysis implemented in R and Matlab, we explore the relationship of average vehicle occupancy statistics, population density, community clustering, and average trip length distributions with residential housing prices in each county of the United States. In this way, we seek to determine whether the potential for residential ride-sharing in a particular county is an approximate predictor of the home prices in the region and whether the value associated with greater ride-sharing opportunity is incorporated into those prices. We also conduct a financial analysis of county real estate markets using machine learning to predict prices over the next 20 years based on the historical performance of real estate markets in relation to the other major financial asset classes. In addition, as smart-driving services like Google’s Waymo have already begun to pick out locations for testing driving capabilities in Phoenix, Arizona, and Detroit, Michigan, this thesis identifies precise counties of the United States that are most suitable for the introduction of ride-sharing technologies to service residential areas and connect them with areas of business and attractions.