National Institute for Transportation and Communities

The video begins at 3:33.

The video begins at 9:25.

This paper, co-authored with Ian W.H. Parry, derives formulas for the welfare effects of reforming subsidies for peak and off-peak urban rail and bus fares, and applies them to the metropolitan areas of Washington, D.C., Los Angeles, and London. The model accounts for congestion, pollution, oil dependence, and accident externalities associated with automobiles and each transit mode. It also accounts for scale economies in transit supply, costs of accessing and waiting for transit service, crowding costs, pre-existing fuel taxes, and the transit agency’s adjustment of frequency, vehicle size, and route network in response to changes in demand. We find that in almost all cases existing subsidies – which typically exceed 50% of operating costs – are either about right, or possibly too low, across bus and rail, peak and off-peak period, in the three cities.

Speaker Biography: Kenneth A. Small, Professor Emeritus of Economics at the University of California at Irvine, specializes in urban, transportation, and environmental economics. Recent research has concentrated on urban highway congestion, measurement of value of time and reliability, effects of fuel efficiency standards, public transit pricing, and the role of fuel taxes in managing external costs of automobiles. Prof. Small served five years as coeditor of the international journal, Urban Studies, and is now Associate...

Watch video

View slides

As metropolitan area governments and others promote density-promoting “smart growth” policies, finer analysis is needed to quantify the impact of such policies on households' transportation and housing costs. Existing research suggests that households in urban areas trade-off between housing costs and transportation costs, but does not explore how policies to increase urban densities might explicitly impact this balance. Furthermore, the research does not adequately distinguish between the effect of urban area density and the effects of other factors associated with urban area density (e.g metropolitan area size and household incomes) on housing costs. This research uses the 2000 Census Public Use Micro Sample (PUMS) person and household data from 23 of the nation's most densely populated states to identify the impact of increased population density on three housing cost measures: household rents, housing unit values, and monthly mortgage payments. Log linear models were estimated for each housing cost measure using least-squares regression. Dependent variables included household, housing unit, and geographic area characteristics, including population density. The models were found to be very similar to one another in terms of the statistical significance...