The Infrastructurist discusses  a recent study that suggests that an increase in gas prices leads to a reduction in sprawl. Here is a summary of the study:

Georges Tanguay and Ian Gingras analyzed data from the 12 largest metropolitan region in Canada for the period of 1986 to 2006 and found that higher gas prices “contributed significantly” to less sprawl:

On average, a 1% increase in gas prices has caused: i) a .32% increase in the population living in the inner city and ii) a 1.28% decrease in low-density housing units…

Tanguay and Gingras addressed this shortcoming by expanding their observations over a 20-year window. The researchers found the aforementioned link between higher gas prices and reductions in sprawl. They also report that a 1 percent increase in gas taxes led to a .2 percent reduction in commuting distance (though the effect is small, amounting to just 14 fewer meters of travel, on average).

The researchers did notice a potential mitigating factor: income. Every 1 percent rise in median income led to a .23 percent decrease in city center living. That means any reduction in sprawl that occurred as a result of rising gas prices could be offset by rising income.

So if gas prices went up more than $2 on average in the US between late 2008 and today (roughly a 140% increase), then we would expect the inner city population to grow by 44.8% (.32% increase in population*140) over the same time period? Perhaps this is extrapolating beyond the scope of this data but this would be quite a population shift. Even a smaller increase in gas prices, say 10%, would lead to a predicted increase of 3.2% in inner city population, still a sizable increase.

It would be helpful to take the same kind of analysis and apply it to American metropolitan areas. Does the same relationship hold? I suspect it might not as some big central cities have not really gained much population in the last decade (see the case of Chicago or New York City). Could some of this observation come from how the Canadian government measures city centers or from a higher proportion of Canadians living in the “city center” (the study suggests the proportion of the population living in city centers is “the average for Canadian CMAs is 55%” – the American population is at least 50% suburban)? Does Canadian culture have less emphasis on sprawl (and single-family homes with yards, driving, etc.) compared to American culture?

This is an interesting finding but I would be interested in seeing more research on this. A 2004 American study cited in the discussion reached this conclusion: “The results show that every penny increase in the state gasoline
tax in the late 1980s is associated with nearly a five square-mile reduction in the size of an average urbanized area.” Additionally, I would be curious to hear more about why this study used the “average-sized” urban area in a state as the dependent variable:

The dependent variable, the average-sized urban area in the state, ranges from a high of337.8 square miles (Arizona, given the large size of the Phoenix metropolitan area) to a low of29.34 square miles (West Virginia). The mean of the dependent variable is just over 120 square miles, which, for point of reference, is slightly more than double the size of the urban area contained in the Burlington, Vermont metropolitan area, or just under the size of the urbanized land area in the Anchorage, Alaska metropolitan area.

I see that the gas tax measure of interest is at the state level but using state level data for cities seems strange as urbanized areas can vary quite a bit (think of the comparison between Chicago, IL and Springfield, IL – both urban areas but quite different in scale and urbanization). Additionally, a measure like the percentage of state residents who use public transportation to get to work would seem to be related to the size of urban areas. Why not simply use each urbanized area as a case?