Source: The fourth power law
One 18 wheeler does the damage of 10k cars.
Freight rail saves on costs and repairs of roads.
Now look at trucks (18 wheelers) and try to decide if they’re cheaper than trains when factoring in infrastructure maintenance.
I was an over the road trucker for a bit, and this was one of the first things that struck me. Going through Chicago is a literal river of trucks 24/7. Absolutely no reason 90%+ couldn’t be a train. Just fucking embarrassing really. We let the money management bros into the train system and this is what we get.
A big part of this is about who pays for the infrastructure. In the US at least, most roads are paid for by the public whilst railways are paid for by the company that owns them. To make matters worse, while the cost of making a 13 lane highway is externalized, many states charge taxes per track mile, which incentivizes single-tracking.
Essentially what you end up with is that if you’re sending goods by train, you’re paying for both the maintenance of the train tracks and the roads the trucks use, whereas if you send them by truck you’re only paying for the road maintenance. This is a direct government policy that selects for trucking over rail, despite the inefficiency.
if you send them by truck you’re only paying for the road maintenance
But you’re not even paying that. You’re only directly paying the vehicle tax on the truck, its fuel and amortized operation and maintenance costs. But the vehicle tax doesn’t even come close to covering the cost of the damage the truck causes to the road infrastructure. You pay the difference indirectly in other taxation that is a subsidy to the trucking industry, and also taxation that subsidizes the fossil-fuel industry.
So the bias against rail transport is even greater than you indicate.
This is why I think large companies with lots of trucking should be paying a lot more taxes for roads and bridges. As it stands now, ordinary citizens are subsidizing them while they turn around and raise prices off the back of this. Corporate welfare for nothing in return
And that leaves out all the other externalities where a car also has overwhelmingly greater negative impact than a bicycle.
According to this math the entire population of the Netherlands has to bike on one single road more than 80 times to create the same amount of damage as one truck…
One truck does the damage of 10,000 cars actually so multiply that times 10,000.
Where bicycles typically ride, it is likely a greater discrepancy.
Unsurprisingly, it’s more complicated than that, but…
-
The literature found a range of damage law exponents from 1 to 12
-
On average, state highways…should consider using a damage law exponent of approximately 2; however, designing for the heaviest commercial vehicles operating on local low-volume roads with a lower life would need to consider a damage law exponent closer to 6.
One consequence is that, even at the lowest exponent, there is absolutely no reason to charge vehicle tax on bikes, if the rationale is to compensate for externalities, rather than just brainlessly punishing cyclists.
-
So what would be the difference between a typical sedan and one of the monstrosity truck & SUV things rolling out lately?
Monster truck with the extra big knobby tires? Or regular production monster truck?
Uuuuhhuhhhhh both
I’m going to guess it’s even more since they rounded up the the nearest tenth. An ebike is closer to 0.04 tonnes. A regular bike is closer to 0.02 tonnes. So probably above 300,000.
I’m pretty sure you need to include the people riding the bikes too, since they weigh more than the bikes.
All while the bus damages the road as bad as a few thousand cars.
True. And buses are far lighter than fully loaded semi-trucks.
Its also only a partial story as “damage done” doesn’t directly relate to actual costs.
Take Ottawa as an example
Transitway is nothing but buses all day long, and that has an amortized annual cost of $42,100 per lane/km/year.
A local road that sees a couple hundred car trips a day costs $14,122 per lane/km/yr.
So that’s 3× the capital cost for way way way more vehicles at 1-3,000x the “damage” per vehicle.
Bicycle lanes an amortized captial cost of around $5-1000 per lane/km/year (this number is REALLY hard to peg down due to all the different ways cities account for bike infrastructure and the type of infrastructure it is).
So a bike lane is somewhere between 14 to 3000x less expensive than a local road, despite 160,000x less “damage”
Yeah, that’s because other things besides vehicles damage roads too; the weather being another big contributor to road damage, especially in places that have cold winters.
It’s sensible policy to understand how each factor contributes to costs (and benefits), but it’s not sensible to assume there are only one or two factors.
Yeah, and that just capital costs.
With the same Ottawa example, a sidewalk isn’t much cheaper than a road in in operating costs, but a pathway is nearly nothing.
That’s based on paths getting little to no snow and ice clearing.
Yes, but it’s pointing out that maybe fixating on a singular metric could backfire.
More wheel surface area probably reduces this somewhat. I suspect that it’s the fourth power of the pressure, with the number of axles being used as a proxy for this.
It’s probably still well over four orders of magnitude, mind you.
The empirical finding that yields the fourth-power law is based on vehicle weight, not pressure.
I’m sure somebody has done far more detailed modeling, but that’d entail consideration not only of weight distribution, but the properties of different road surfaces and their relative frequencies of occurrence relative to road usage patterns. Modeling all that can get messy fast. Hence the populatiry of the fourth-power rule of thumb, which isn’t a bad gross approximation.
Imagine three cars:
- Car 1 is a Honda Civic. Perfectly ordinary, just like you find all over today.
- Car 2 is the same model of Civic, but modified so that it has eight half-sized tires, four per axle. It has the same contact area with the ground and the same mass. It’s pretty intuitive that this would not significantly change the amount of road wear.
- Car 3 has been modified relative to car 2 so that it has four axles with two wheels per axle instead of two axles with four wheels each. Same mass, same contact area, just distributed a bit differently. Can I prove that car 3 doesn’t cause 1/16 the damage of car 2? No, but I’d be very surprised if it did.
Yes, axle weight is a reasonable proxy, I don’t disagree. However, when making broad statements, it’s good to be as precise as possible.
