Matthieu de Lapparent

Risk and Uncertainty in Urban and Transport Economics

Urban and transportation economics often require discrete decision making in risky situations. Inoa, Picard, and Palma, in the article ‘‘Effect of an Accessibility Measure in a Model for Choice of Residential Location, Workplace, and Type of Employment,’’ develop discrete choice models to understand residential and professional choices. They study the best way for residents to acquire information in a sequential choice, with an application to Île-deFrance. Xin and Levinson, in their article ‘‘Stochastic Congestion and Pricing Model with Endogenous Departure Time Selection and Heterogeneous Travelers,’’ describe the choice of departure time with risky travel times and simulate its effects on transportation policies. They use the standard dynamic model for this purpose. Cohen, in her article ‘‘Risk Perception, Risk Attitude, and Decision: A Rank-Dependent Analysis,’’ reviews the literature on decision making under risk and uncertainty either with expected or with non-expected utility. She shows how the expected utility model is extended in the context of nonexpected utility theory. Inoa, Picard, and Palma combine discrete choice theory with the modeling of choice in risky alternatives. The decision process is hierarchical. In practice, when the decision maker faces many alternatives, usual economic models are inoperative and both the decision maker and the modeler must simplify and regroup different alternatives concerning residential location choice and employment choices. The authors analyze a three-level decision process: residential location, workplace, and type of employment. If decision is sequential, the information available ex ante, at the first step, is different from the information available ex post, once the first step choice is made.

Risk Aversion in Travel Mode Choice with Rank-Dependent Utility

Using 2004 stated preference data on travel mode collected in the Zürich area, different parametric specifications of the rank-dependent utility function in a logit mixture model show that commuters are weakly averse to small-time losses. The results also justify Yaaris dual theory of choice under risk, that the utility function is linear on outcomes but that the perception of corresponding probabilities is biased. For leisure travel, the travelers are risk neutral to small losses of time.

Risky Time Prospects and Travel Demand

Every traveler makes choices in a stochastic environment because of unpredictable disturbances coming from both exogenous incidents (e.g., weather conditions) and aggregate endogenous behavior of other travelers (e.g., accidents, unexpected congestion). These nonrecurring disruptions to a traffic system result in variable travel times. However, a structurally assembled and robust modeling of demand and supply under risk is far from being achieved in transportation analysis. It should explicitly be accounted for as it affects related choices of modes, routes, and travel schedules (Palma et al., 2008). Appraisal of transportation policies would be more accurate if considering not only the value of travel time savings but also the value of travel time reliability (Bates et al., 2001; Palma and Picard, 2006, 2010). Assuming that travel choices are made under risk modifies the modeling of demand and the analysis of the distribution of traffic flows on transportation networks. Despite contributions (Abdel-Aty et al., 1995; Noland and Small, 1995; Noland et al., 1998; Bates et al., 2001; Small et al., 2005; Brownstone and Small, 2005; Avineri and Prashker, 2005; Palma and Picard, 2005, 2006; Chorus et al., 2006; Lapparent, 2010; Ben-Elia, 2011), there is no general consensus on choice behavior under risk because of competing theories, lack of appropriate data, and testable empirical models. The three contributions on modeling travel choices under risk presented in this special issue reflect the current state of the art and extend usual models. The authors focus on mode choice, route choice, and departure time choice.