Removing an elevated city highway doesn’t always make traffic worse—some cars just disappear.
Today the Toronto City Council will debate the fate of the Gardiner Expressway—a mile-long stretch of elevated road that cuts through the city. Local sentiments seem roughly split at the moment. The lead city planner, who’d like to replace the segment with a surface boulevard, has pushed for the Gardiner’s removal. The mayor is backing a “hybrid” option, essentially keeping the structure in place, for fear that commuters will suffer too much in its absence.
[UPDATE: The council voted 24 to 21 for the “hybrid” option, though those on the removal side said they’d continue to fight.]
It hasn’t exactly been a peaceful exchange. Here’s Mayor John Tory’s response when asked whether the chief planner had been “restricted” from discussing her opposing position:
The only thing, generally, I think public servants should not be doing is sort of debating politicians because they are public servants and there is a line to be drawn there.
The debate over removing urban highways remains a familiar one around the world. (It’s an especially hot topic in U.S. cities, where elevated urban highways built as part of the interstate system near the end of their functional lives.) In the case of Gardiner, and most of the others, the sticking point comes down to traffic.
Consultants seem to favor removal.
Dillon Consulting, which has taken the lead on the Gardiner project for the city, punts to the city council in its conclusion about whether or not the expressway should come down. But if you read its report carefully, the consultant seems to favor removal in all but explicit name.
On Dillon’s scorecard of the options, “remove” is preferred in the areas of urban design (a teardown “allows the full development of a urban district introduced by a tree canopied urban boulevard”), the environment (a teardown gives “lower green house gas emissions and greater opportunity to create new natural habitat”), and economics (a teardown has a “lower net 100 year lifecycle cost”). The only area where hybrid wins out is transportation, primarily “due to the lower auto travel time.” Of the 16 total sub-measures that Dillon scores, the remove option is preferred on 11.
On cost, where politicians usually fall, remove is the overwhelming preference. The numbers here reported by the media vary, but just considering the lifecycle cost of the infrastructure itself, the Dillon report shows that the hybrid option will reach $919 million (in 2013 dollars)—whereas the teardown will come to $461 million. When you consider that a number of developers have lined up to reimagine the waterfront area in the absence of the elevated expressway, the numbers fall even more heavily toward removal.
How much are three minutes of delay worth?
The only real question, then, is whether or not the costs of congestion that might emerge by tearing down the highway will exceed the economic benefits that seem in store for the Gardiner area. It’s a fair one to ask. There’s an enormous amount of productivity and personal time lost to traffic in heavily congested cities, and the steady flow of goods in and out can be vital for both business and quality of life.
Dillon estimates that removing the expressway will create an extra 2 to 3 minutes of delay during the morning rush-hour over the hybrid option come 2031. That doesn’t seem like much on paper, and it feels even less significant when considering the small share of commuters who rely on the Gardiner to get into work. Dillon estimates it at 5,200 cars an hour—or just 3 percent of all morning commute trips.
Mayor Tory doesn’t consider this delay a trivial price, and has been quoted as saying the per-trip delay could be as bad as 10 minutes. That figure would raise some alarms, but it’s a very questionable one—coming from a study commissioned by several trucking associations. The study itself doesn’t emphasize the 10-minute delay possibility, offering that as an outer range of estimates that are as little as zero extra minutes per trip (Slide 45).
And even if the study’s figures are to be believed, it estimates a top annual congestion cost of $36.6 million—well shy of bridging the cost gap between the two options.
But drivers adapt very quickly when a road is removed.
Every individual case is different, and the reality is that many urban interstates carry far too many cars to seriously consider total removal. But generally speaking, drivers find a way to adapt when a road is removed—many cars simply go away. As New York City traffic guru Sam Schwartz told a Toronto Globe and Mail reporter, in reference to the teardown of an elevated highway in Manhattan decades back, “there was this disappearing effect that we could not quite trace.”
This “disappearing traffic” phenomenon has been studied carefully, and it seems to be a pretty consistent one. A 2002 paper, which looked at 70 case studies of reallocating road space from general traffic cars to other modes, reported a median traffic reduction of 11 percent and an overall average decline of 22 percent:
The findings suggest that predictions of traffic problems are often unnecessarily alarmist, and that, given appropriate local circumstances, significant reductions in overall traffic levels can occur, with people making a far wider range of behavioural responses than has traditionally been assumed.
Writing at the blog Architect This City about the Gardiner debate, transport planner Darren Davis of Auckland, New Zealand, gives a simple four-part explanation for what happens to traffic when a major roadway gets removed or altered. Some people change their routes. Some shift their travel times to hit the road earlier or later. Some switch from cars onto public transit or another mode. And some—typically the case for non-work travel—just don’t make the trip at all.
At the end of his recent op-ed favoring a teardown, Globe and Mail architecture critic Alex Bozikovic asked: “But are we building a city? Or a highway?” Toronto won’t live or die based on the Gardiner decision, but it will be answering that critical question.