Theoretical and empirical investigations of excessive congestion as a result of market and planning failures

This dissertation research places traffic congestion in a broader context of land use and economic linkages and contends that optimal congestion relief requires both land use and pricing policies. Congestion is considered excessive when the individually desirable (or privately optimal) amount of auto travel exceeds the socially optimal level. Two underlying causes of excessive congestion are discussed in detail here: market failures from congestion and agglomeration externalities and planning failures from exclusionary zoning and low-density zoning practices. This research is among the first to connect the economics of planning failure with excessive congestion. This research first specifies a spatial general equilibrium framework that reflects congestion delays, agglomeration economies, and planning failures. Simulation findings suggest that anti-congestion policies might erode agglomeration economies, causing a net social loss. Pricing policies need to balance the benefits from congestion reduction with the losses that come from weakening agglomeration tendencies. The congestion diseconomy is only a small share (5%-23%) of the total cost of congestion; policies seeking to produce free-flow speeds may lead to substantial welfare loss. Simulations demonstrate how that, when planning failures dominate region, even the first-best pricing is not so effective since planning failures are insensitive to (market) pricing signals. Incorporating land use and economic policies are found to be socially optimal when both planning and market failures exist. This research also examines practical policies. Application of a mileage tax or a cordon toll partially reduces excessive congestion and decentralizes jobs. Urban growth boundaries are relatively inefficient and may distort land markets, causing worse congestion. Firm cluster zoning is more effective since allowing for jobs decentralization. Densification policies can alleviate the excessive congestion caused by low-density zoning regulation. Under exclusionary zoning regulations, building an employment subcenter can greatly improve welfare and alleviate congestion. Planning for new subcenters, however, requires a subsidy or incentive to trigger the firms’ relocations. This research also presents an empirical study using the 2006 Household Travel Survey data obtained for Austin, Texas. This study develops a multilevel multinomial logit model to investigate the interaction effects between land use and travel cost variables on travel mode choice. Results suggest that road-pricing policies are more efficient in reducing driving in neighborhoods with better walkability and easier access to activity centers. The impacts of land use patterns on driving are stronger when driving costs rise. These findings suggest that an incorporation of both land use policies and road pricing policies benefits a region’s residents more than the either policy on its own.