As a general rule, road pricing isn’t popular. But that’s not because it doesn’t work. The problem is that it works too well, and people don’t like the idea of driving less and paying for roads (that currently have a zero marginal cost).
Here’s a recent study by Robert Bain and Deny Sullivan that looked at just how well it can work. In it, they examine 76 data points from 16 countries, including roads, bridges, tunnels, and cordons (areas).
The question: What happens to demand once the marginal cost of using a road goes from $0 to some cost greater than zero? (As part of this, they also looked at whether the road or bridge in question has viable alternatives.)
The results:
The median traffic reduction was 25%. But the interquartile range was -17% to -44%. This is all very significant. Said differently, the traffic impact in nearly a quarter of the examples was -45% or more. So almost a halving of traffic congestion.
These reductions are obviously a function of the cost of using each road, but regardless, the overarching takeaway remains the same: You may not like or want road pricing, but it totally works.
For next year's budget (2024), the City of Toronto is projecting a $1.5 - $1.7 billion budget shortfall. And over the next 10 years, this shortfall is expected to grow to nearly $47 billion if changes aren't made. This is according to a recent report prepared by Ernst & Young and Strategy Corp. So right now, all of this is being looked at and debated by Council.
Where are we going to get this money?
One persistent debate is whether the city actually has a revenue problem, or whether it's simply an expense/spending problem. I can't say that I've scrutinized the city's expenses at any length, so I'm not going to get into that level of detail today. For this post, I'd like to focus on two specific things. The first is property taxes.
Here is a figure, from the report, showing residential property tax rates across southern Ontario:
What you will see is that Toronto has the lowest rate of the 35 municipalities that they looked at. Now obviously there are some nuances to consider. The average home price in Toronto is higher than it is in, say, Sault St. Marie. Toronto also has a large commercial property tax base. But even still, historically speaking, Toronto has tended to increase its residential property taxes at or below the rate of inflation.
This is a problem. And it is the exact same problem that we have talked about on this blog in regards to residential rent controls. If you own an apartment building where the rents are capped and your expenses are, therefore, growing faster than your revenue, you are (1) highly incentivized not to invest in the apartment (you can't afford to) and (2) eventually going to hit a financial wall.
Sound familiar? As far as I can tell, that is, at least partially, what is happening here.
Secondly, one of the first things that I did when I opened the report was run a search for "road tolls" and "congestion charges". Regular readers of this blog will know that this is something I feel strongly about. Here's what I found:
In 2017, when the City considered implementation of tolls for the Gardiner and the DVP, staff estimated that a $2-per-trip toll would generate $5.6 billion in 10 years. The province has refused several requests to consider these options, with the Minister of Transportation rejecting any discussion of uploading or tolling as recently as December 2022.
This is also a problem. One of the general rules with taxes is that you should ideally tax the things you want less of. Hmm. So why not tax traffic congestion? There is no question that it works. There's lots of evidence from all around the world. We just lack the political will to actually do it. Instead, we pay lip service with solutions that don't work.
At the same time, if we were to actually implement road pricing, I don't believe that a flat toll is the way to go. $2 also seems low. The best practice is dynamic road pricing that fluctuates based on actual congestion levels. Meaning, if you're driving at 5am, expect a low rate. And if you're driving at 5pm, expect a high rate.
Virtually overnight, we know this would do at least three things: (1) it would reduce/eliminate traffic congestion (congestion levels would become a function of pricing); (2) it would reduce overall carbon emissions in the city; and (3) it would take a meaningful chunk out of this $47 billion budget shortfall.
Daniel Knowles, who is a correspondent for the Economist, recently authored a book called Carmageddon: How Cars Make Life Worse and What to Do About It. I haven't read it, yet, but I did just read this excerpt about Tokyo, and it was jam-packed with interesting stats.
Here are some of them:
Among developed cities, Tokyo has the lowest car use in the world. About 12% of trips are completed with a car, whereas 17% of trips are done with a bicycle. Most people walk and/or take transit. Tokyo has the most-used public transit system in the world -- about 30 million people each day.
Car ownership across Japan is about 590 vehicles per 1,000 people. This is comparable to many European countries. In the US, it's about 800 vehicles per 1,000. However, this figure drops in Tokyo. Here, the average is about 0.32 cars per household, which was interesting to see because most new housing projects in downtown Toronto have parking ratios that are much lower than even this figure.
The average size of a home in Tokyo is 65.9 square meters of usable area. By comparison, the average size of a home in London is 80 square meters. But given that according to Knowles, the average household size in London is 2.7 people, whereas it's 1.95 in Tokyo. So per capita, Tokyoites actually have more space than Londoners.
35% of streets in Japan are not wide enough to fit a car. If you add in streets that are wide enough to fit a car but not wide enough that a car could stop and not entirely block traffic, this figure jumps to 86%. This to me is a massively significant statistic, because if you want people to walk places you need small streets.
95% of streets in Japan do not allow any sort of street parking -- day or night.
The average Japanese car owner drives around 6,000 kilometers per year. This is about a third of what the average American does. In my case, it looks like I have averaged about 8,868 kilometers per year over the last 5 years. Though a big chunk of my kilometers would be from longer one-off snowboarding trips. In other words, I don't drive all that often in the city.
Japan has some of the most expensive road tolls/prices in the world. Meaning, Japan does not actively subsidize driving and instead just charges drivers accordingly. Apparently the average is about 3,000 yen per 100 kilometers, which is about CA$30 per 100 kilometers.
In addition to not subsidizing cars, Tokyo is also one of the few cities in the world where their public transit does not need to be subsidized. A big part of this has to do with high ridership, but the other important part is that its transit authorities also develop real estate. Shockingly, this means that it tends not to build standalone and single-storey transit stations (ahem, I'm looking at you Crosstown LRT). Instead, they build lots of density where it always belongs: on top of transit.
I may just have to read Knowles' book.