Here’s an insight from tolling:  A substantial portion of the people driving on our roadways are only there because we’re subsidizing the cost of their trip.

When we charge a toll to use a road, suddenly many of those using it find they don’t value it enough to pay even a fraction of the cost of the road’s cost.

Our most egregious subsidies are to those drivers who demand to travel at the peak hour when roadway space is scarce, and which is the most expensive problem to fix.

The biggest problem with transportation is we don’t price road use correctly, to reflect back to road users the costs that their decisions impose on society and everyone else traveling.  In essence, we have daily urban traffic jams for the same reason Ben and Jerry have long lines on their annual Free Cone Day: when you don’t price something valuable, it gets rationed by queuing and patience.  What’s worse is a lot of people who are driving on many roads at the rush hour is essentially because we are paying them to do so.

Freeways are only “free” in the sense that, no matter when you drive, you pay the same marginal price:  Nothing.  Of course,  we all pay for roads through things like the gas tax and vehicle registration fees (as well as a host of other general taxes), but the point is, whether you use a road at the peak hour or not has no effect on how much you pay to drive.  Whether you use it at 2 am when there are literally no other cars on the road, or you use it when it’s jammed at rush hour, makes no difference.  And because peak hour capacity is in short supply, and is damned expensive to increase, “free” roads are a massive subsidy from the general public to the relatively few of us who use roads at the peak.

When we ask road users to pay even a modest fraction of the cost of providing that expensive peak hour road capacity, many of them tell us, by voting with their feet (or their tires), that they simply don’t value the roadway enough to pay for even a tiny fraction of its cost.

The latest example of this comes from Seattle, where Washington State has spent about $3.3 billion tearing down the old elevated Alaskan Way highway viaduct that marred the city’s waterfront for decades, and replaced it with an expensive new tunnel bored under downtown.  After the tunnel was completed, it was “free” for a while, but a year ago, the state asked users to pay tolls, which vary from $1.25 to $2.25 per trip, plus a $2 surcharge, if you don’t have a transponder.

Now keep in mind that tolling will cover less than 10 percent of the cost of the tunnel, about $200 million of the more than $3 billion cost. When the tunnel was first opened, it carried more than 75,000 cars per day.

The Alaskan Way Viaduct (Wikimedia Commons)

As soon as the state asked those using the tunnel to make a modest contribution to its cost, that number dropped to 55,000.  Here’s the official analysis from WSDOT:

Of the 20,000 trips that diverted from the tunnel after tolling began, data indicates that roughly 10,000 trips used other routes (primarily Alaskan Way, Elliott Avenue, and I-5) or switched from driving to riding transit. Ferries and water taxi ridership were largely unchanged. The remaining 10,000 trips are not accounted for using existing roadway sensors and automated passenger counters on buses, however, anecdotal data suggests that some trips were discontinued as people embraced teleworking, bicycling and other active forms of transportation, or by using an unmonitored route.

What this really means is more than a quarter (20,000 of 75,000 users) used the tunnel because only somebody else paid for more than ninety percent of the cost of their trip.  And contrary to the usual doomsday scenario’s of highway agencies, most of that traffic isn’t displaced onto city streets or alternate routes, it simply disappears.  Again, the evidence from Seattle’s tunnel has been that neither abrupt changes in capacity or tolling produced noticeable worsening of traffic elsewhere in the city.

Here’s another example, also taken from the State of Washington, although it spills over—as we’ll see—into the state of Oregon.  To set the scene: Vancouver Washington, sits just north of the Columbia River, which forms the border between the two states, and is part of the Portland Metropolitan area. Washington has a sales tax, Oregon doesn’t.  There just two bridges—I-5 and I-205—connecting the 400,000 Vancouver area residents with jobs and stores in Oregon.  Washington residents who shop in Oregon avoid state and local sales taxes of more than 8 percent, meaning a $200 shopping trip saves a resident more than $16 in sales taxes.  Washington is, in effect, paying its residents to drive to Oregon to shop.

Washington residents fill the parking lots at Oregon stores to avoid their state’s eight percent sales tax; they’re essentially paid to drive to Oregon to shop.

We’ve estimated that the average Vancouver household saves more than $1,000 in sales taxes per year by shopping in Oregon and that shopping trips account for between 10 percent and 20 percent of the traffic on the two Interstate bridges.  Not incidentally, this shopping traffic is a key reason why the two states are considering spending more than $3 billion to widen the I-5 bridge.

Want less traffic and pollution? Stop paying people to drive

This has to be the key insight for transportation policy:  The reason we have traffic congestion is essentially because we’re paying people to drive.  As soon as we ask drivers to pay even a fraction of the cost of the roads they’re using, large numbers of them find other routes, or simply take fewer trips.  In a world where we’re losing the fight against climate change principally because we’re driving more and more each year, a logical first step is to stop paying people to drive their cars.  That’s exactly what policies like road pricing, and parking pricing do:  they ask those who benefit from using their cars to pay for a larger fraction of the cost of providing the services they enjoy.

One of the fascinating, and least widely understood aspects of the science of traffic jams is that it takes only a small reduction in traffic levels to keep roads flowing freely.  As long as traffic levels stay below a tipping point where the road becomes saturated and loses capacity, the road works well. Adding (or subtracting) just a few cars when traffic is at this tipping point makes a huge difference both in travel times and how many cars a road can carry. This is just what happened during the height of the pandemic, when reduced demand kept Portland-area freeways below their tipping point, and enabled them to carry more cars at the rush hour than they did in “normal” times. Pricing a roadway causes a minority of travelers to change their trips, but it’s enough to keep the roadway flowing freely. A common argument against pricing is that some people simply don’t have alternatives to driving, or driving at a particular time; and that’s true.  But the data show that a significant minority of those on the road will immediately change their behavior given even a small financial incentive—and that segment of the population staying away is enough to make the road system work much, much better for everyone else who doesn’t have that flexibility.

The old saying “you get what you pay for” applies with a vengeance to transportation:  We pay people to drive, and so they do, with the result that we get chronic congestion, air pollution, and aggravated climate change.  A sensible policy of charging road users based on when and where they travel would both dramatically reduce traffic congestion—by encouraging those who are only on the road because someone else is paying for the trip to choose another time or destination—and also reduce pollution and greenhouse gases.  Those who paid would be getting what the British call “value for money”—by tolls would get them a quicker and more predictable trip than is possible at any price now.