Comment by Out_of_Characte | Hacker News Reader

Comment by Out_of_Characte | Hacker News Reader

Out_of_Characte

This article does not make sense to me. City's main point of attractiveness is high density. When you have a place with lots of resources ( usually near rivers ) people will move there. The more people, the more business which means if there's any place for any nice market then hitting the biggest cities first is a good bet. Yes, this means the more houses you build in an area, the more people will move there which will increase the attractiveness of a city. Attractiveness can be represented as value. This, in no way means you should stop building houses in places that are the most attractive. Quite the opposite. Build until the price drops or the cost of even higher density outstrips the benefit that could be gained. Eventually too many people will be priced out. You might not like the idea that only the very rich can occupy the highest value spot in a city but thats just free market.

"They added two more lanes to the freeway, but the traffic is worse than ever.” But that’s a wonderful result"

No, its not. Thoughput is a function of speed * vehicles. Having a 6 lane highway at a crawl is as effective as a single lane highway at decent speed. Induced demand is complicated to grasp but the core is that highways are terrible at moving lots of people and any additional highway will make more people drive to match the highway's new capacity. A neverending traffic loop. Trains, trams, buses and cyclepaths all move more people which should be prioritised. Many countries atleast have priority for buses which makes them a better option. Most american cities treat them like another car but with lower status.

> any additional highway will make more people drive to match the highway's new capacity.

If you build a 6 lane highway in Alaska, it's not going to get traffic. Induced demand is really supressed demand; people avoid trips because the cost in time is too high, increasing throughput allows more trips to be made at the same time cost, and so more trips are made.

But, if you look at Caltrans reports before and after highway widening, there's usually a decrease in the hours of congestion. At least until population growth overcomes the increased capacity. Mass transit can certainly move more people per vehicle and that can make more throughput, but it also limits the viable trips. If you loom at somewhere like the LA area, it's hard to live near where you work, especially if you have two workers in the household or you change jobs. Some routes have mass transit options, but a lot of commutes would take all day without a car; especially if the commute is at off hours.

The myth of induced demand needs to die a short and painless death. The confusion between something at capacity being expanded and the expansion itself causing the increase is too prevalent.

Okay then figure out another term to explain why bumping Katy Freeway up to 28 lanes is not going to fix Houston's traffic problem.

The real explanation is more subtle which is that sure, you can just keep adding freeway until there's no traffic, but it's actually bad to have monster 30+ lane highways running through the middle of your cities.

It's not a cute name, but you can't measure actual demand when you're at capacity --- maybe hidden demand, unmeasured/unmeasurable demand. But population growth is often ignored too --- if your metro is growing rapidly, and your road expansion project takes several years and traffic 'is the same' after construction, that's actually a successful project; if the expansion didn't happen, traffic would most likely be much worse.

Think of a webserver with a 100M link; if it's at capacity all day long, you know you should upgrade the link, but should you add a second link so you get 2x100M, or does it need 1G or 10G or more? Hard to say. If your 100M link is only at capacity for a few minutes or even an hour a day, adding another link is likely to be enough. Of course, network capacity is easier to add than road capacity.

I understand all this, the only bit relevant to the point is “network capacity is easier to add than road capacity.”

That is why people disparagingly call unmet demand for roads “induced demand,” because it makes it clear it will be very difficult to just more-roadway your way to low traffic.

Ok, the problem with the nomenclature is it sounds like the demand for trips just magically appear when the road does. If you don't want to satisfy the demand for trips with additional road capacity, that's fine; additional road capacity is expensive, but then so are most forms of additional trip capacity, but using a misleading term to defend the choice is not a great way to do it.

You probably wouldn't be happy saying we shouldn't build additional housing in (wherever is housing capacity limited) because it will just induce demand for housing and it will still be capacity limited. Certainly, you'd need to build a lot of housing in San Francisco or New York or lots of metros to satisfy the unmet demand for housing and unless/until you build enough, housing will still be a congested resource, but there's no inducement; same thing with roads.

At some point, if you're not going to build housing and you're not going to build transportation (whether that's roads or rails or bike paths or flying carpets or whatever), you need to have a real coming to terms of this metro is full and how do we manage its population in a reasonable and fair manner.

Does public transport not create induced demand? Won’t a new tram line encourage people to make journeys that weren’t feasible before?

Exactly. That's why tram+bike combo in Netherlands is so effective. People travel big distance by train and small distance to work, home and shops. Since trams are very efficient at moving a lot of people car congestion will be reduced. If number of people using trams increases more than their capacity, surprise, you can add another section without hiring one more driver. So big capacity transport that can be adapted to demand and that has priority on the road(it's own lane) and is frequent enough basically can satisfy most of the traffic human needs

The scaling curve of a tram or bus is very different than for cars. A bus takes up about as much room as 3 cars and holds up to 200 passengers. The typical car at rush hour has 1 passenger.

You can pack a lot of buses onto a bus lane. Trains and trams scale even better because you can expand the same vehicle by adding more sections.

200 passengers in the space of 3 cars? Lol

An articulated bus can officially carry 120 passengers[0]. With regards to road space used, 3 SUVs with normal following distance would be approximately the length of the bus, or longer. An articulated bus is about 19m, while an SUV is about 4.5m.[1]

[0] https://en.m.wikipedia.org/wiki/99_B-Line#:~:text=It%20is%20....

[1] https://en.m.wikipedia.org/wiki/Nova_Bus_LFS

The official capacity of the busses in my city is 52 seated and 94 standing. They frequently go way over capacity during rush hour, so around 180 passengers definitely isn't unreasonable. While not quite as much as the person you were replying to claimed, it still illustrates the point.

This could be, but usually public transport is more effective. Eg. A Bus can transport 4-8 more people, than a car for the same space used on the streets. This rate is even more increased with trains or trams. Therefore people need to demand 4-8 times more public transport than with cars to equalize the effects.

This holds true as long as the population is constant, becaus a single human can not spend an indefintly amount of time in public transport.

If the population can increase eg. Housing Markets in cities, this wont work. If you could double or quadruple the amount of available houses instantly the demand would also go up after some time…

Yes, the main benefit of trains and proper buses is that you wont get stuck in traffic. So in the worst case scenario you'll see troughput increases as a function of ridership (more occupancy, more trains) while trying to battle overcrowding. By adding more trains and buses. Trains have another benefit to the population that they can travel at higher speed, so the induced demand with trains can get even higher because of the large distance you can travel if you travel at 200+ kmph.

Induced demand isn't actually a bad thing unless your transportation gets less efficient the more people use them. Which is definitely the case with cars. Just look at the amount of parking spaces needed to cope with peak hours. This can all be avoided if trains and buses are a viable and comfortable option for people.

Just think about the average car ridership only having 1.2 passengers. While a bus that takes up three parking spaces can have 50-100 people. Or a train that could hold up to 1500 passengers.

Reply on news.ycombinator.com