A challenge question for PT readers:

How should we start to limit congestion before it becomes unacceptable?

There’s a real-life experiment unfolding on our streets – one that will fundamentally affect our future – discussed here in “Our Real World Experiment in Traffic Congestion. “

As people switch from transit to cars, it won’t take much to fill up available road space.  It may only take a 10-15 percent to reach a level of inefficiency and frustration where we reverse the gains we’ve made in this region, notably with transit, in the last half century.   Without response, something has to break, even if we don’t yet know what that level is.  Waiting until we get to a breaking point seems kinda stupid knowing how much more difficult it is to reverse something if instead we can limit it before it happens.

Knowing we will have to slow, stop and reverse a move to post-Covid motordom worse than pre-March, what steps should we take now?

I say we should start by treating congestion the way we do the weather: agree on what we will measure and how we will do it continually and consistently. Produce really slick graphics to make it understandable, provide regular reports – especially forecasts – and give names to the ugly stuff.

Engineers and planners know we measure what matters, and they are good at doing it scientifically.  But that’s not enough.  Every day we have to tell the ongoing story of our changing transportation patterns at this pandemic-shaping time.  It’s essential if citizens and leaders are willing to go along with drastic actions – just as we have learned from the way we measure, report and proscribe our response to the virus itself.

Here’s an action plan:

(1) Choose key points to measure how fast congestion is emerging – most likely the bridges, but also locations along suburban and exurban roads – and display with graphics similar to weather reports.

(2) Report on these measures as they’re happening – hourly if not in real time. and provide forecasts as we do with weather, air quality, even Covid-19. Use highway signs, as with variable speed limits, to inform drivers on the road.  After all, Google already gives us real-time indicators of congestion.

This is not a substitute for the A-F classification system, based on a false premise that anything less than the posted speed limit is a measure of congestion.  This is telling the story of system failure when we depart from the strategies that actually work – notably transit and active transportation design to accommodate growth, not by building more motordom infrastructure that worsens the problem.

Then the most important step:

(3) Set the limits beyond which we do not wish to pass as a social target – the points at which congestion inefficiencies become too much.  Sure, “too much” is subjective, but it won’t take long before people can see the relationship between the reported data and their experiences on the road.  They then need to know at what rate it is going to get worse, and how unfortunate that would be.

We’re more likely to mobilize and respond as a society when we know the nature of the problem, a measure of its reality, and the choices we need to make in response that have and will work.  We’ve already done that, haven’t we, with Covid.  Now we need to do it with the problem created by the changes in the way we move.



  1. If governments had the discipline not to intervene, traffic capacity would limit itself. The best thing to do is to do nothing. Most drivers finds their own saturation points – the limit at which they simply won’t tolerate sitting idle in their cars. When that happens they find religion and do what humans always do – they adjust. They switch to transit, drive off-peak, work from home as often as possible, or switch homes or jobs to suit. Economists frame this as the point at which the cost of driving (the perceived ability to drive “freely”) equals its perceived benefits. Drivers don’t tolerate ever-rising costs for no additional benefit. At some point, peak hour driving becomes intolerable and people adjust. Vancouver had the foresight to realize this decades ago and has thankfully maintained its discipline.

    Let us also remember that what we think of as “congestion” (a term that perpetuates the problem, like reflexively calling a multi-fatal vehicle crash an “accident”) is peak only. We don’t need to do anything to gauge the effects o f this on the road network during these crappy COVID times. Motorists will do that themselves. The total daily volumes could surpass pre-COVID traffic and it wouldn’t necessarily matter. What matters is the peaks. If those compressed volumes return, people will adjust. Only now they have a bit more leeway to drive off-peak than they once did.

    The hope is that city governments maintain the discipline to do nothing. Don’t widen roads because some drivers piss and wail about slow traffic. Don’t remove pedestrian crossings or bike lanes to gift motorists even more space because they “need” it. Just let drivers find their new equalibrium. Their levels of tolerance for traffic have not changed. When those limits have been reached as they were previously, they will adjust – if we let them.

  2. I’m not sure that we need to do anything differently than we did before. Covid doesn’t change anything. Sure it caused a temporary dip, but that dip already seems historical, and likely things will return to normal next year.

    The three options are:

    Build more roads. These increases capacity for a bit until growth and induced demand fill up the space and move the constriction point to other points in the system. But, besides induced demand, it also leads to a restructuring of life and commerce to be more auto dependant with more distant shopping, work and recreation locations. Things like Silvercity Riverport and Tsawwassen Mills. When that happens it really loads the system and usually overwhelms any previous traffic improvement (and damages the city in other ways). Or increasing capacity can just accommodate growth in the existing pattern without leading to a more auto dependant pattern. Whether or not such growth is a good thing is a question, but we ought to acknowledge that it is a possibility because it is a big reason why the public support road construction. And if the public do want that growth, other measures must be taken to avoid establishing a more auto dependant pattern and the other possible degradations that more traffic can bring to life.

    Provide other options. These are usually transit, walk and cycling improvements that such that the benefit of those options are greater than the benefit of driving is. (Or the cost of those options are lower than the cost of driving, depending on your point of view.) Or they are urban form options that require less auto use to live in. Here in the region we have done well with providing transit options, assuming we are just in a lull before embarking on another construction cycle, and we have done OK on improving cycling options. The City of Vancouver is much better at cycling infrastructure but other municipalities aren’t really interested in it. The urban form front is also quite mixed. There are definitely the germs of better urban villages in many areas but most are not fully formed. And they are neglecting two key ingredients for walkable cores: small blocks and narrow facades, ie, don’t have one building occupy the whole block. Unless it is a very interesting building. Like the Opera Garnier. I haven’t seen auto and biking data at a fine enough scale to know how well these urban villages or town centres are performing save for the ones on skytrain routes that have higher transit usage.

    The third method is to increase the cost of road use. We have a gas tax that notionally funds roads or notionally charges for the environmental damage, but that isn’t directly linked to particular roads or particular times of day when the congestion cost that you impose on others varies greatly. A regional road tolling scheme of fine enough grain to charge everyone for road use is what is needed. Everyone would have an rfid chip on the car, probably incorporated into the licence plate, and there would be rfid readers embedded in the road. We could start with readers on all the bridges and highway entrances and exits and expand the system such that everyone travelling more than one km would be sure to hit one. The gas tax would remain to charge to the environmental damage to fossil fuel use, but all the other taxes that fund transportation ought to be rolled into the toll. Eventually the toll would be merged with the fare such that a transportation user would be charged for the use of public goods, be it labour, infrastructure or land (which is most important public good involved in roads). Needless to say, this is the option that I prefer.

    There is of course the do nothing option which lets congestion be self limiting. I’m not in favour of this option. First, I don’t think that it is politically realistic. We have continued to expand and build new roads in this region when official policy have been to limit such endeavours and we have also in some circumstances moved toward a more auto dependent urban pattern because of them. See Tsawwassen Mills. If new roads are built, better to have within them a self limiter. Second, congestion has an economic impact on commercial traffic. Construction workers and repairers won’t come to your house if you live in a hard to get to part of town. But the third reason is that I really don’t like the status quo. Congestion stuffs the city with cars, particularly around choke points. Those choke points might help some areas stay clear, but they are not pleasant for those choking at the choke point. A reasonable goal ought to be to have fewer cars on the road at any given time.

Leave a Reply

Your email address will not be published. Required fields are marked *