UC Irvine Institute of Transportation Studies

Some road-pricing demonstrations use an approach called "value pricing", in which travelers can choose between a free but congested roadway and a priced roadway. Recent research has uncovered a potentially serious problem for such demonstrations: in certain models, second-best tolls are far lower than those typically charged, and the welfare gains from profit maximization are small or even negative. That research, however, assumes that all travelers are identical and it therefore neglects the benefits of product differentiation, by which people with different values of time can choose a suitable cost/quality combination. Using a model with two user groups, we find that accounting for heterogeneity in value of time is important in evaluating constrained policies, and improves the relative performance of policies that offer differential prices. Nevertheless, for most of the reasonable range of heterogeneity, second-best pricing produces far fewer benefits than pricing both roadways optimally, and profit-maximizing tolls are so high that overall welfare is reduced from the no-toll baseline.