University of California Transportation Center

This dissertation examines why and how jurisdictions choose to finance theirroads. The systematic causes ofrevenue choice are explored qualitatively by examining the history of turn pikes. The question is approached analytically by employing game theory tomodel revenue choice on a long road. The road is covered by a series of jurisdictions seeking to maximize local welfare. Jurisdictions are responsible for building andmaintaining the local network. Complexity arises because local network usersmay not be localresidents, and localresidentsmay use non-local networks. Key factors posited to explain the choice of revenue mechanism include the length oftrips using the road, the size ofthe governing jurisdiction, the degree of excludability, and the transaction costs oftoll collection. These factors dictate the size and scope of the free rider problem. It is hypothesized thatsmaller jurisdictions and lower collection costsfavortolling policies overtaxes. The analyticalmodel is operationalized by assuming jurisdictions have two decisions: the strategic decision to tax ortoll, and the tactical decision ofsetting the rate of tax ortoll. Models of user demand as a function oftrip distance and monetary cost and of network costs as a function oftraffic flow and the number oftoll collections are specified. The values ofthe constants and coefficients ofthemodel are developed fromrecent cost literature and the estimation of amodel of collection costs from California Toll Bridge data. The model is applied to evaluate the dissertation’s hypotheses. The application evaluatesthe welfare implications of a jurisdiction and its neighborsimposing general tax, cordon toll, odometertax, or perfect toll policies. Sensitivity tests ofthemodel under alternative behavioral assumptions, and with varyingmodel coefficients are conducted. Finally, policy implicationsfromthe analysis are drawn.The general trends which bode well forroad pricing (electronic toll collection (ETC), decentralization, advanced infrastructure, privatization, and federalrules) are established. Possible scenariosforthree cases are presented: deploying ETC and building new tollroads, and converting free roads.