This thesis uses applied econometrics and traffic experiments to identify environmental and behavioral factors that contribute to externalities in traffic networks, as well as evaluate mechanisms designed to address them.

The first chapter examines whether exposure to ambient fine particulate matter (PM 2.5) increases the likelihood of getting into a vehicle collision. PM 2.5 has been shown to affect alertness and cognition, which may in turn impair driving ability. Variation in daily AQI level from PM 2.5 was exploited to identify a possible causal effect on daily car accident rates in nearby cities. This approach yielded no evidence of a causal effect on vehicle accidents, perhaps due to endogeneity of PM 2.5 with other factors correlated with accident frequency. An alternative instrumental variables approach exploited exogenous shifts in wind direction relative to nearby coal power plants - a significant point source of PM 2.5. This specification found that a one-standard deviation in PM 2.5 AQI increases the car accident rate by 13.2 percent.

The second chapter investigates if the presence of multiple states in traffic networks adversely impacts the speed at which users learn route-choice equilibria. To answer this question, data were generated from several sessions of a repeated binary route-choice experiment with human subjects. Exogenous random state changes were introduced as discrete, varied reductions in roadway capacity. The sessions were comprised of either a “simple” network treatment with only two states, or a “complex” network treatment with five states. Reinforcement learning models estimated from the experimental data show that learning was significantly impaired in the complex five-state treatment but not the simple two-state treatment. Simulations based on the learning behavior estimated from each treatment showed that the impaired learning from the five-state treatment resulted in disproportionately slower (and sometimes non-existent) equilibrium convergence compared to learning with two-states.

This third chapter demonstrates the workability of a truth-telling mechanism for efficiently allocating freeway capacity. I conduct a traffic experiment using an interactive multi-user driving simulator in which I allocate human subject drivers to freeway lanes using an optimal tolling scheme where users reveal their valuation of the road through a Vickrey-Clarke-Groves mechanism. I find that the mechanism generally elicits truthful values of time from subjects. However, there are also significant and persistent deviations from truth-telling caused largely by difficulty in understanding the complexity of the mechanism as well as stochasticity in travel time outcomes. Nevertheless, I show that the mechanism dominates alternatives under a plausible set of assumptions.

Travel demand forecasting models play an important role in guiding policy, planning, and design of transportation systems. There is no shortage of literature critiquing the accuracy of model forecasts (see, for example, Pickrell, 1989; Wachs, 1990; Pickrell, 1992; Flyvbjerg, Skamris Holm, and Buhl 2005; Richmond, 2005; Flyvbjerg, 2007; Bain, 2009; Parthasarathi and Levinson, 2010; Welde and Odeck, 2011; Hartgen, 2013; Nicolaisen and Driscoll, 2014; Schmitt, 2016; Odeck and Welde, 2017, and Voulgaris, 2019), not to mention several high-profile lawsuits (Saulwick 2014, Stacey 2015, Rubin 2018). Many researchers and practitioners feel more can be done to advance rigorous travel analysis methods for the public good (see, e.g., zephyrtransport.org). Motivated by these critiques, a two-day, NSF-funded workshop was held at UC Berkeley in the Spring of 2017 to engage in a fundamental review of the state of the art in travel demand modeling, to discuss the future of the field, and to propose new directions and processes for advancing the science.

Travel demand forecasting is an inherently practical enterprise. While academics drive the fundamental research, the users of travel demand models and forecasts are typically government agencies and transport operators that use the models to inform long-range investment, funding, and planning decisions. Private firms play a key role in assisting the agencies in both development and application of the models, and, more recently, high-tech firms have entered the development fray. While all of these actors have important roles in advancing the science of the field, in this report we focus our attention primarily on the academic side of the enterprise, consistent with the orientation of the funding agency (NSF), and in order to make the task manageable. That said, other sectors are represented in various parts of this report as they interface with academics or play particularly central roles in our proposals for advancing the science.