With oil over $100 a barrel (as of early March 2008), and gas prices averaging 70 cents/gallon more than what they were a year ago, bloggers and media outlets are abuzz with the idea of "the end of sprawl". The idea is fairly straightforward: according to the basic model of urban economics, households and businesses trade off accessibility (proximity to a central business district) for floor space. One can havelow transportation costs and high real estate costs by locating near the urban core, or low real estate costs and high transportation costs by locating in the suburbs. With higher transportation costs all around, each household or business will choose to live closer to the urban core, increasing density. How plausible is this story?
The impact of increasing fuel prices on urban density is not obvious. First, the trade-off between accessibility and floor space could just lead to higher land prices near the urban core. Second, even a doubling of gasoline prices may not raise total transportation costs that much, when total costs take time and car ownership into account. Third, higher gas prices could increase consumer demand for more fuel-efficient cars. Let's look at each of these points in turn. We will see that the "end of sprawl" story has elements of truth, but the total effect of increasing fuel prices may be small.

Lessons from Urban Economics

How are housing prices determined? Assume that households have a fixed amount of money to spend on housing and commuting. A household that lives in the suburbs pays a higher commuting cost for access to the central city. That household is therefore willing to pay less for housing, and housing (and land) prices in the suburbs are accordingly lower. What happens if transportation costs go up? That depends on what your other options are. If the household can only move within the metropolitan area, the prediction of a shift from suburban to city living might be correct. Households have the same budget as before, so they will try to reduce transportation costs by moving closer to the city center. More people trying to live closer to the center will drive up central city housing prices. Most likely the household will have to reduce other consumption in order to make up for an overall increase in combined transportation and housing costs. What if households can move to a different metropolitan area? If transportation costs increase in one place while staying the same in other metropolitan areas, then the increase in the cost of living will cause some households to depart for greener pastures. The place with higher transportation costs will shrink in population and spatial extent, but the remaining population will be less dense. Of course, this means more population and higher housing prices in the low transportation cost cities. Which is the best approximation of reality? If transportation costs increase everywhere, there is no advantage to moving to a new metropolitan area, and the closed-city model is the appropriate one. But an increase in oil prices could lead to an increase in the cost of car commuting while having no impact on the cost of electricity-powered public transportation such as New York City's subway and (most of its) commuter rail lines. One possible outcome is an increase in the population of transit-oriented cities at the expense of car-dependent cities.

How Much Does Fuel Cost Matter in Total Transportation Cost?

The urban economic analysis tells us what happens if transportation costs increase, but how much will rising fuel prices increase commuting costs? Gas is only part of the cost of owning a car. The Edmunds.com website allows you to find the true cost of ownership for a variety of cars. The Toyota Camry, America's top-selling sedan, has a five-year true cost of ownership (including depreciation, insurance, taxes, maintenance, etc.) of ~$40,000, of which fuel costs are ~$10,000 (assuming 15,000 miles driven per year). The Chevrolet Suburban (I couldn't resist), a large SUV, has a five-year true cost of ownership of ~$63,000, of which fuel costs are ~$15,000. In each case, fuel costs represent about a quarter of the cost of owning a car. If gas prices were to rise 50% (to near $5/gallon), this would raise the cost of owning a Camry by about $1000 per year, about $1500 per year for the Suburban. While not insignificant, compared to the total cost of car ownership the rise in fuel prices would most likely lead to relatively small adjustments in vehicle miles traveled (VMT) or household location. And that's leaving out another significant part of transportation cost, which is people's time. Empirical studies have estimated that people value their time at about half their wage. The average commute in 2006 was 25 minutes (US Census Bureau Tables B08012 and B08013), which, assuming a five-day workweek translates to ~250 hours/year spent commuting. Assuming average employment earnings of $20/hour (slightly higher than the Bureau of Labor Statistics national 2005 estimate of $18.62), the average person "spends" about $2500/year in time costs of commuting. If we add this time cost to the total cost of ownership for our hypothetical Camry owner, fuel costs now account for only 19% of the cost of commuting. Again, adding in increased fuel prices could lead to some reduction in VMT or a shift toward living closer to one's job, but not a radical break from past patterns of suburbanization.

The Fuel Efficiency Response

Finally, we have been examining consumer response to an increase in transportation costs, but an increase in fuel prices does not necessarily translate into an increase in transportation costs. In response to the Oil Crisis of the 1970s, consumers demanded more fuel-efficient cars and the federal government introduced fuel economy standards. It's not cheap gasoline that encourages people to live farther from their jobs, it's cheap transportation. Transportation can be inexpensive in two ways: first, the monetary outlay could be low; second, faster travel times reduces the time cost of commuting. In fact, it could be argued that fuel economy standards increase sprawl by decreasing transportation costs!