Hacker News Reader

Comment by lazyjones | Hacker News Reader

This nonsense gets repeated over and over again:

What you get is what you build. New roads just create more incentive for driving. It's been well documented over and over again that vehicle miles driven increases proportionally to roadway created. Lewis Mumford famously said “Building more roads to prevent congestion is like a fat man loosening his belt to prevent obesity.”

It's not the roads that cause traffic, it's the convenience of driving. To show how absurd the argument is, let's use it for housing - I'm sure it can be easily "documented" that building housing leads to higher population figures. So it must be bad, let's not do it!

View on HN

You guys are saying the same thing. More roads make driving more convenient than other modes, increasing the incentive.

And yes, increasing housing supply lowers the price, and increases population. That's the point; to move people into the core who want to be there but otherwise couldn't afford it.

Taking away roads doesn’t make cars any less convenient. It just means that fewer people will travel to that area.

Mission success I guess?

Well, it actually does.

The convenience of a car is determined by how likely you choose a car over other means when you need to get somewhere. If there's less roads, you're less likely to choose car, thus the lowering the convenience factor. If there's no roads at all, owning a car is meaningless since you can't drive anywhere, and its convenience is pretty much zero.

So you've still made the whole experience less convenient in absolute terms but made another option more convenient relatively. That is not a success. It's like dumping a pile of venomous snakes in the upstairs bathtub so now I have to go downstairs, thus now the downstairs bathtub is more convenient.

I'm not arguing about that and not trying to call it a success. I'm just stating that taking away roads does in fact make cars less convenient.

Roads make driving more convenient at the expense of other transportation methods; being able to walk to the store is more convenient that driving there. You can't do it though if there's this huge highway in the way and your city is so low density that the store is miles away.

If this argument doesn't convince you, think also that driving may be convenient, but I find more convenient if we lived in a planet that's not a literal oven

I wouldn't find walking to a store more convenient than driving unless the walk to the store was less than the walk from the parking lot at a store I'd drive to.

even if it takes less time? for driving you have to get to the car, drive, deal with traffic, park, and so on

Driving for 10 minutes is generally more appealing to me than 5 minutes of walking carrying bags, but it depends on the situation. I bike a few minutes to a nearby McDondalds sometimes but then often drive that same distance.

Sure, if there are no roads at all.

But if SF makes a street pedestrian only, it just means that area is more inconvenient. And it might not even mean that, since the inconvenience will reduce demand so those who still drive there may have about the same level of convenience.

Replacing roads with public transit does not mean that fewer people will travel there. It often means more, as proven in hundreds of cities around the world.

> It's not the roads that cause traffic, it's the convenience of driving.

Exactly. Roads make driving more convenient.

Build trains and that makes taking the train more convenient so you get more of that.

Seems pretty straightforward to me.

The problem I have with mass transit is that outside of a couple major cities, the system is a horrible time waster. I used to take a bus to downtown job in my mid-sized Midwest city and it was like commuting 2x--I would have to drive to a park and ride, then wait for a bus, then board a bus and then go sit in traffic on a bus. I would arrive 40 mins later, dropped off near the building I worked at, but still had to walk outdoors in sub-zero temps and deal with low-level street crime at times, too. Yes, I got to read or do other crap on the bus, but all the colds and other sickness I caught didn't make up for that.

Contrast that with me just getting in my car and driving to the downtown parking garage and paying for parking at a place that has a skyway where my feet never touch the street at all, and you can understand the dilemma.

Then I got really smart and just stopped working downtown completely and got a job in the suburbs with free parking and while the traffic is still bad, it's nothing like heading downtown is.

One thing to note here is that roads and driving have been heavily subsidized while mass transit has been fought against.

Mass transit is usually subsidised too. Very few mass transit systems survive on fares.

> Then I got really smart and just stopped working downtown completely and got a job in the suburbs with free parking and while the traffic is still bad, it's nothing like heading downtown is.

In my case, working downtown would have been the smart move. I lived in a suburb and worked in a neighboring suburb. The drive was 10 minutes and I could avoid the freeway. I looked into taking the bus because there was a bus stop right outside my apartments and found it would take a minimum of 40 minutes, but realistically with traffic and decent timing on the transfer, it would have taken over an hour. The reason? All the bus routes were essentially straight shots to and from the city, so to take the bus I'd have had to go all the way into downtown, switch buses, and ride all the way back out.

> sub-zero temps and… colds and other sickness

Look into a vitamin D3 supplement, approx ~5000 IU daily in the wintertime. Haven't been sick more than a day in the last 5-10 years.

Convenience is only part of the story though. Price matters - and for me at least it is often as cheap or cheaper as a single person in a car to drive than get a train. I live in London, I can train anywhere pretty conveniently. But, somehow, if I decide that I want to go to Leeds next weekend to see my friend - it's £100.

Well, shit. The train is more convenient - but that's over twice the cost of petrol. Now imagine that I wanted to take another friend with me. It's absurd. The same scenario, but to a friend in Rugby (about half as far, different train line) costs me £23. Bargain!

There are multiple axes on which to nudge behaviour - and one of them is definitely price.

The trains in the UK are an outlier here. You're not wrong, it's outrageously expensive — it was cheaper for me to rent a car for 24 hours and drive from London to Bristol and back than it was to take the train. Between London and Newcastle it's sometimes cheaper to fly (and I don't even mean with a budget airline!). This is utterly insane. It doesn't need to be this way though. Trains work great in continental Europe. In fact the trains here in Poland are fantastic.

I'm a total petrolhead and I'm often irritated by the kind of cyclist that hates all cars and anyone who drives one, but I would like to see fewer cars on the road. Trains are an excellent way to achieve that goal (just not the way the UK does it).

This reminds me of Germany, where flights between Munich and Berlin are at a constant 40€(easyjet)/50€(Lufthansa) for booking next day, since the public train company has a 4hr service during "rush-hour" now (nextday booking costs 150€ without a discount card - which is 250€ for a 50% rebate for a year)

It's only cheaper because you're not paying for how much it will cost to get the combusted petrol back out of the air again. You're stealing that from future generations.

Trains don’t solve the last mile problem in metropolitan areas. It’s just not economically feasible to constantly run trains or buses that cover a metropolitan area with less than 1 mile spreads and acceptable intervals.

I’m being generous that ‘walk a mile’ at source and destination would pass muster for people with difficulties.

Making our infrastructure for cars better seems like a much better solution honestly.

Walk a mile at source and destination also adds 30 to 40 minutes to your trip.

Taxis, shuttles, rideshare are a thing.

Right, but the point is that all the solutions to the last mile problem are... cars. Which is fine but if you're going to do that then what's the advantage of not using it the whole way? The throughput of highways is great, there's no schedule to follow, you don't have congestion and surge pricing at popular stops with hundreds of people trying to hail a cab at rush hour, and if you're price conscious gas is cheap and you can find perfectly usable cars for less than $1k.

And you can use your car for more than just your daily commute.

Parking is the main difference.

It seems like a perfect opportunity to let the free market solve the last mile problem.

I mean the market has been solving the all mile problem with cars for the last 50 years. Outside of the downtown areas of ~5 cities in the US there isn't really a problem to solve.

So the article goes on to suggest expanding a bus network or other public transportation. In this manner, it increases the convenience of a more traffic-efficient bus and decreases the convenience of the less traffic-efficient car.

We're unfortunately in a place where public transit systems are starting to fall apart in many major US cities (I dont know if its all), such as Boston and NYC. At the same time, we're trying to reduce car usage. That's a pretty hard pill to swallow for many people.

If it was a cohesive strategy, where housing and transportation plans are moving together, there would be a lot less friction.

> It's not the roads that cause traffic, it's the convenience of driving.

That's a tautology, roads are a convenience for driving.

Busses and commercial traffic also use and require roads. They don’t exist just for those commuting by car.

It's easy to prove that statement false just by getting outside of big cities. I'm sure if they expanded I-5 in Weed, CA from 4 lanes to 6 you'd see zero impact on traffic.

I don't think you disagree. If new roads make driving more convenient, doesn't it stand to reason that building new roads also increases the number of miles driven?

Induced demand is not nonsense.

I'm willing to believe that, but it always sounds like nonsense to me. Normally, I hear people use it to describe the basic economics of a demand already greatly in excess of supply. If there's not enough food to go around, providing a little more food just means people will eat that additional food and they will still be hungry. We don't conclude that induced hunger is a thing.

Going back to traffic as the example, if I want to drive downtown to see a ballet but decide against it because traffic is bad, then they add more lanes and next year I do decide to drive to see that ballet, then I'm creating more traffic because they added more lanes. That demand wasn't induced -- it just transitioned from unmet demand to met demand.

If induced demand isn't nonsense, then I must be oversimplifying it. What am I missing?

> We don't conclude that induced hunger is a thing.

Certain foods are more expensive than others. That's why we don't eat steak and lobster for every meal. But if the price of those foods went down, you can bet there would be a lot more demand. That's induced demand. Everyone was already getting enough calories and nutrients, but making something desirable a lot cheaper had the effect of more people trying to buy it.

> But if the price of those foods went down, you can bet there would be a lot more demand.

No, you can bet there'd be a greater market-clearing quantity. Greater demand would mean people were willing to buy more at thesame price. If anything, over time, lower prices should reduce demand by setting s lower expectation of the “right” price and reducing the quantity people are willing to buy at higher prices.

Demand is the function mapping price to quantity people are willing to buy.

Yep. And the government fixes the price of using a road at $0, causing the induced demand issue. A limiting factor of supply (how many cars are currently using the road) is completely ignored when setting the price.

The cost of constructing roads is handled by other taxes, but that doesn't change that the roads themselves need to be priced higher to not have shortages.

> And the government fixes the price of using a road at $0,

That's arguably true of electric vehicles, but ICE engines pay road user fees by way of gas taxes, which are nonzero and related (imperfectly, sure) to road impact.

> causing the induced demand issue.

I think induced demand is not really a road price issue: more road access -> more desirability to attractive businesses -> more demand (at the same price) for road use.

Now, if you charged higher road prices instead of building more roads, to reduce traffic, you'd avoid induced demand by avoiding attracting business and raise revenue, it's true, but the induced demand isn't a product of $0 road cost (which isn't a thing, mostly) but of increased ease of access from the short-term effect of more roads.

I wasn't arguing that induced demand is a product of $0 road cost. It's a product of lower time cost - new roads decrease congestion (temporarily) and induce the higher demand. Eventually the new roads are as congested as the old ones.

Induced demand is unmet demand. Rural towns have no problem getting head of demand by building more. It is only in dense cities that nobody builds enough roads to meet the unmet demand for roads.

Of course I didn't mention cost above. A rural area can meet its demand with a cheap asphalt two lane road. A city would need dozens of levels of bridges at a much higher prize to get ahead of demand. Which is why cities should look to transit which is not cost effective for rural areas.

If it's nonsense, care to provide some supporting links? The author provided 4 links, one of which was from the California DOT supporting the statement.

Downs-Thomson Paradox is useful to cite here:

https://en.wikipedia.org/wiki/Downs%E2%80%93Thomson_paradox

These are dubious theories from decades ago, if taken literally, they'd mean we'd just need to remove all transport (public included) and there would be zero demand for them. Behind such absurd claims there's always a simplistic model with wrong assumptions and too few variables. Thankfully, new roads are still getting built despite every attempt to refer to these bogus paradoxes.

> It's not the roads that cause traffic, it's the convenience of driving.

But more roads increase the convenience of driving, no?

Your example is poor. Housing may lead to a higher local population, but no one copulates based on the national availability of housing.

Anyway no ones mentioned induced demand, which is the relevant term

https://en.m.wikipedia.org/wiki/Induced_demand

Vast majority of people don't drive just because a road got wider either. Build a 2x3 highway pointing nowhere and you'll get just a dozen of enthusiasts hold an illegal race on an empty road...

No, but your commute is probably bounded by time, not distance. If you can only travel 5 miles in X time, you'll get the best job in that radius. If you can travel 60 miles, the odds are your job will end up being further away. Rinse and repeat for shopping, socialising, etc, etc, etc.

Cool. Now you have more job opportunities, can keep up with your friends further away and have wider choice of shopping/eating out/etc place. No need to stick to shitty place next door.

really cool

Actually, that nowhere place then becomes possible to develop. And that development will increase traffic, on the new road and on the roads it connects to.

That's the real reason you build more and wider roads. But it's for some reason sold as a way to reduce traffic.

On this side of the world usually it's vice versa. Developers start building on cheap land with little infrastructure. Then increasing population forces government to improve infrastructure decades later. By the time construction is done, it's already outdated and overcrowded.

Even if it happens your way... Is it wrong that an interesting location opens up at the end of new road? E.g. a restaurant or trailhead? Yes, it would increase traffic. But on the other hand, this would also let people do things.

> no one copulates based on the national availability of housing[...]

The GP's point is nonsense, but people most definitely do start families based on the overall cost of having kids, and housing plays a major role.

Again though I would say house prices are locally decided, not nationally. San Francisco house prices aren't influenced by $cheap_area_across_country

Reply on news.ycombinator.com