Hacker News Reader

Comment by JimboOmega | Hacker News Reader

Is this even a problem? People are given an opportunity to do something and then they do it. That's way better than new roads going unused, isn't it?

I like congestion charging alright, which the article advocates at the very end, but it has a habit of being regressive - it's easy for me, as a software engineer, to show up at work at 11AM, but that might not be the case for those who can less afford to pay those charges. I think it's a net positive, but it can have consequences - especially in places like the Bay Area where there is no vaguely affordable housing anywhere near where the jobs are.

Why not think about locating things people want to go to - especially their jobs - near where they live? Why do we need sprawling suburban "campuses" and a million tech busses to shuttle people from the cities to them? Throwing up huge office buildings and tech HQs in SF - but refusing to build housing the same way - is only a little tiny bit better.

Both just force people to the exurbs, creates more super commuters and more demand on the roads.

View on HN

It is, yeah. Roads are bad and toxic to the businesses and areas around them. Nobody wants to hang out next to a 4 lane blacktop with cards screaming down it at 60 miles an hour.

It spreads out businesses and makes our cities utilities spending less efficient. The roads need to be maintained, and it is very expensive to maintain them.

https://www.strongtowns.org/journal/2018/1/16/why-walkable-s...

https://www.strongtowns.org/the-growth-ponzi-scheme

> Is this even a problem? People are given an opportunity to do something and then they do it. That's way better than new roads going unused, isn't it?

It's a problem insofar as building more roads is often sold as a way to resolve traffic problems. When traffic expands to fill all available roads, this solution is an expensive way to not fix anything.

There are probably more useful options, and faster ways to waste money, available to polities.

But it is making trips possible that weren't before; it's not strictly money going down the drain. I have trouble seeing a fully utilized resource as "a waste".

Saying "traffic will be solved by this!" is clearly misleading. But those extra cars on the road must be benefitting, or else they wouldn't make the trip.

That's an excellent and very wise question! It's one that deserves careful study. At which point different ways of accomplishing the same goals can be compared and the best option(s) selected. I think we can agree this would be a reasonable, level-headed, approach to make informed decisions.

This may be slightly different from a political cycle that spends ever-increasing amounts of resources on roads without examining if the results obtained provide a reasonable level of value for money. Or indeed address the issues that drove the construction of additional roads in the first place.

Broadly, fully utilized resources are not always an ideal goal. For example, my experience is that meeting rooms in offices are almost always in constant use, no matter how many there are. This may not be the same as all uses of meeting rooms being desirable and efficient uses of the time and other resources used. I know I have been in meetings, enabled by a sizable of supply of meeting rooms, that could have been handled much faster with a handful of emails.

Fully utilizing the resource means some benefit is obtained, certainly. But when the cost of the resource to the user is zero, that means the benefit only need be greater than zero.

Have you ever worked on a shared compute cluster? In general, disks fill to capacity & machines run at capacity, no matter what those capacities are. It's bloat; files that don't get deleted until you run out of disk, because deleting files takes ten seconds. Massive processes that are left always-running to save ten seconds of launch time. So on and so forth. Basically, the user is getting some value. But it's very small, and thus makes a poor reason to make expensive upgrades to the cluster.

> When traffic expands to fill all available roads, this solution is an expensive way to not fix anything.

It always surprises me to see people actually believe this utter illogical nonsense. There's a finite amount of people and cars, so new roads satisfy existing demands but do not magically increase traffic until everybody drives 24/7 and beyond.

There are plenty of roads with very little traffic. Hasn't anyone of those anti-car-activists noticed?

Of course it’s true that people and cars are finite, but that doesn’t mean that new roads are satisfying existing demands. In fact, the term in economics is “induced demand.” Building more roads around a city often won’t decrease traffic in the city, it will just make more people commute to the city from further away.

> In fact, the term in economics is “induced demand.” Building more roads around a city often won’t decrease traffic in the city, it will just make more people commute to the city from further away.

It's a theoretical concept with no basis in reality as far as traffic is concerned. Every statistic I know of shows that more people move to cities, not farther away despite new roads. What drives people to longer commutes is not new roads, but high housing prices in cities.

Also, when you build useful roads, traffic there certainly increases, but somehow ideologists neglect to measure how traffic on other roads previously used for the same route decreases.

I’m not so sure about that. The Wikipedia article on induced demand has no shortage of cited studies.

> Is this even a problem?

It’s a problem if your assumption was that demand was closer to fixed and building more roads would “spread drivers out”, reducing congestion.

This was for decades, and still to some extent is, a very wide-spread assumption in local politics, leading to incredibly questionable urban planning decisions.

If we paved every square inch, traffic wouldn't go to infinity. It's pretty rare to go driving for no reason. Country roads don't have rush hour.

It seems reasonable to assume that latent demand per capita is fixed, but at a level too high for a city to accommodate without destroying its density advantages.

People will also adjust their housing and working decisions based on road availability, although the lag here is a little long.

If the center is already built and occupied to its legal limit, you can't get a population-level change from this.

But people will move to the periphery and commute farther, or change where their job is.

I don’t think anyone proposed people would drive for no reason. When there is more road capacity, people find more reasons to drive. More people get cars and more businesses use cars to move goods around. In the future self driving cars can be used on the road for any business reason someone could think of. There’s probably reason for 10x the cars we have today if they could fit on the road, which most cities could not handle.

In my city they're removing roads left, right, and centre, and using the argument that this will reduce congestion, which admittedly is really bad here. However what's actually happened is people in the suburbans like myself who don't have to good public transportation into the city have effectively become isolated from the city hub. There is really no good way for me to get into the city at rush hour anymore unless I want to spend 2 hours in traffic there and back. There has been minimal investment in public transportation as an alternative, and there has been no attempt to more evenly distribute commercial activity throughout the city.

The problem always seemed to me to be less about road capacity, and more about city density and poor planning. I think we need to rethink the central urban hub model for cities. I suspect small hubs throughout the city would dramatically reduce congestion. We face similar scalability problems in tech. Typically if you have too much traffic going to your server you won't upgrade the server, but spin up more nodes. We distribute the traffic instead of spending larges amounts of money on infrastructure to send it all to a single node.

And another problem of this is that it causes feedback loops. I'm now in a position where I either have to continue to find working outside my city or relocate closer to the city centre contributing more to the urbanisation and population density problem.

With the right government incentives I see no reason why we could reestablish the city centre as a consumer hub. A place for shopping and leisure, etc. Then outside the city centre we could have tech hubs, financial hubs, industrial hubs, etc, all with good (perhaps even free) transportation between them. The time and money this would save seems like a no brainer to me. But then again, I don't really know anything. Perhaps there is a good reason to continue densely packing people into small 2-3 mile radiuses, where space is so limited most people can't afford a car, a decent size home, or even a decent sized office to work out of.

I live in Cambridge about a mile walk from Harvard Square, so fairly close to a decent sized city area, yet we almost never go there. Real estate in the Harvard Square area is so expensive and there are so many tourists that almost all the establishments there are priced out of reasonability for regular/daily patronage. We take out of town guests there and that's about it.

I don't see city center commerce and leisure hubs as being particularly able to dominate neighborhood and suburban equivalents. I need a car to get out to work anyway (8 miles away from the city center), so once all the fixed costs of the car are covered, it's incredibly cheaper and more convenient for us to shop and eat in the suburbs.

I don't see government incentives changing that in a meaningful way.

Government incentives probably won't change the life-styles of people such as yourself who can afford to live near the city core. The problem is for those of us who can't afford to do so, but need to work in the city centre (like most people these days do) our options are basically buy a car and waste hours in traffic everyday, or rent an "apartment" - although these days that tends to mean a small bedroom - in the city centre which you can hardly afford and gives you no satisfaction, nor room to do things like raise a family or have a functional relationship.

What I'd advocate for is governments to tax the hell out of companies who want to buy offices in city cores without a good reason. All a lot of these companies do is contribute to congestion and unaffordable property prices through increased urbanisation which forces people into the city or to deal with the congestion. A few years ago my girlfriend use to have to travel an hour every day to get to the city centre to work in a call centre. What's the point in that?

If instead governments allocated certain areas of cities which have low congestion and offered companies tax breaks for opening offices there you probably would see many moving out of the cities to the suburbs, reducing congestion and making properties more affordable.

I honestly think one of the main factors contributing to depress and mental-illness today, and why people don't have kids like they used to is because of the way we're living. Who the hell is going to raise a kid on £35,000 in 10m x 10m one-bed "flat" in a city like London? However, if on that money living in the suburbs of a city like Birmingham you'd be able to afford a reasonable two-bed home with a driveway and garden. The problem is uncontrolled urbanisation is driving businesses to city centres, and often to the major cities, then forcing anyone who wants to decent job to follow along whether they can afford to or not.

I bet most well paid developers living in the UK here live in London. And that's by no accident. It's not that they all just love London so much that they're happy to live in tiny apartments with no hope of raising a family or affording a mortgage. They do it because that's where all the decent tech jobs are. And why? Who is benefiting from this? A few very rich people and large businesses who can afford offices in city centres?

>> Who the hell is going to raise a kid on £35,000 in 10m x 10m one-bed "flat" in a city like London?

Not sure whether you made a mistake in conversion, but 100 sqm is actually a fairly decent living space here in a city like London (Tokyo).

My 90-odd sqm house has been plenty large enough to raise a family. Small garden, no driveway. But who needs car ownersip when there's three well-connected rail lines within a 7-minute walk?

Density need not be the problem you imagine.

> In my city they're removing roads left, right, and centre

This is fantastic to hear, what forward thinking city do you live in?

Traffic can’t go to infinity, it can only go to 100%, and yes, if you paved every square inch, then traffic would go to 100% in areas where lots of people want to be, like people driving into urban centers for work.

It's a problem because it's a positive feedback loop. More highways = more traffic, which leads people to want to build more highways to supposedly ease congestion, and this continues ad infinitum. Our cities grow more spread out, less walkable, and less sustainable. Overall it makes our cities worse both for the people who live there and the environment.

If we didn't subsidise highways so much, demand for them would be reduced, and it would make more sense to build denser housing and provide public transportation.

Is this even a problem? People are given an opportunity to do something and then they do it. That's way better than new roads going unused, isn't it?

I've never sat in stop-and-go traffic, looked around and thought "Gee, this is great utilization of this road!"

Oddly, I have looked around a crowded standing-room-only train and thought "Gee, these load factors must be good for farebox recovery"

Personally I’d rather be in traffic. I’ll grant trains are way better for the atmosphere and take up less space, but people are noisy and smelly.

People keep saying they don't want to ride on a train or bus because people are smelly, but with rare exceptions, I just haven't run into that problem.

My solution to noisy transit is to wear headphones.

I think the problem is more cultural conditioning, I've been riding transit for 25+ years, and I like taking transit to work. Others who haven't ridden transit don't understand why I'd ride a bus to work even though we get "free" parking.

On the way to work the bus is usually faster than driving (the bus takes the carpool lane), on the way home, traffic is bad enough that the bus is probably about the same speed as driving (maybe a bit longer since for one stop it needs to leave the freeway express lane so needs to cut through 4 lanes of traffic to get to the stop and then again to get back into the express lane), but I don't care since I'm usually reading.

> people are noisy and smelly.

Indeed. And trains usually don't go from people's homes to their destination, so they typically cause traffic of some sort around train stations.

Ideally, people live near transit (not so common in this country, but Transit Oriented Development near transit stations is becoming more popular), so they don't necessarily need to drive to the station.

But even when they do, it's easier to handle traffic spread among dozens of stations spread out over 100 miles of track than to have all of those commuters drive to the city center.

> But even when they do, it's easier to handle traffic spread among dozens of stations spread out over 100 miles of track than to have all of those commuters drive to the city center.

Commuters don't all drive to the city center, they drive to 1000s of working places and arguably that's easier to handle traffic-wise than the traffic around fewer train stations (in one city!).

No, then you end up in a situation like the SF Bay Peninsula where businesses are spread among small office parks that are hard to serve by transit, and it's impossible for families to live near where they work since the spouses may end up working in offices 30 miles apart, so at least one of them is in for a long commute and since there's no planning or zoning that kept their employer near transit, then they end up driving... or taking the employer sponsored shuttle bus if they are lucky enough to have that option.

If businesses were spread out near major transit lines, that'd be another matter, but as it is now, the density is too low to serve well, transfers are a hard sell to commuters, no one wants to commute 40 minutes by train only to have to wait 15 minutes to transfer to a bus that takes another 15 minutes to get them to work.

VTA transit tries to serve a huge area with light rail, but everything is so spread out and there are so many stops that it takes forever, you can take light rail the 12 miles from downtown San Jose to Mountain View, but it'll take over an hour with 28 (!) stops in between. You could bike faster.

That’s not an uncommon view, I’m sure, but I suspect if the externalities of driving an automobile with a single person in it were internalized, so each driver paid the full cost of driving, you and a lot of people would change your minds and instead invest in some (comparably vastly cheaper) noise-canceling headphones for the train/bus.

Minus air pollutuon, I’m not convinced. With air pollution cars are obviously way worse. For commuting your point stands: I prefer current trains to Bay Area traffic because they’re not crowded. But for everything else the car wins by a mile.

The land use required to merely own cars is not great but it’s not the primary cause of sprawl in the US. There are plenty of examples that show car-oriented development with transit to a city center can work at up to 2000 people per km^2 and a typical American suburb has less than a fourth of that population density.

The traffic deaths can be reduced dramatically by reducing miles travelled and speed, but if we’re playing that game they also rank lower than infectious disease which spreads just fine on transit. Traffic deaths also are only a bit higher than suicide in America and lower than it in Europe; suicide is a crude example of the excess death burden of unhappiness, which ought to be considered.

And furthermore I’m not even convinced you’ve been on a crowded train. BART is almost never truly crowded. Trains in Wuhan are crowded. Crowded often means literally shoving your way in if you want to go anywhere. And the person I responded to claimed he thinks about the tollbox operator on a crowded train.

So yeah, dying in traffic on an ugly highway? Cool story bro.

I prefer current trains to Bay Area traffic because they’re not crowded. But for everything else the car wins by a mile.

Which Bay Area trains are you taking that are not crowded? Both BART and Caltrain are full during commute hours -- Caltrain express trains are usually standing room only (in rail cars that aren't well designed for standing passengers) and BART is running crush-loads out of Embarcadero in the evening.

The Caltrain local trains are usually less full but those all stop trains take forever to go any significant distance. (SF to RWC is about twice as long on a local as on a baby-bullet)

Sure, but cars take up 50x more space, create way more pollution, and are responsible for 40,000 deaths in the US a year. So you should not have that option.

I have spent years exclusively riding public transit and very rarely have I encountered an issue where another person became a nuisance. I would rather take public transit than drive any day of the week, both for safety reasons and because I can do other things while on the ride.

People are given an opportunity to do something and then they do it.

If you don't put a price on it then people won't prioritise, and the demand becomes endless. And you can never have supply to meet an endless demand.

It still costs $0.10-0.20 to drive a marginal mile, so the demand is far from unpriced/infinite.

Driving miles anywhere costs the same. Compared to the annual cost of owning a car making an extra 15 mile trip is absolute peanuts. The problem occurs when 1000 people decide to make that very extra trip just because capacity makes it possible and it's otherwise free to use.

> it's easy for me, as a software engineer, to show up at work at 11AM

That's the point -- we should incentivize people who can time shift to do so thereby alleviating traffic for those who can't.

We should incentivize not driving as much.

What’s the point in me commuting to work to write code on a couch in an office that feels like my living room anyway?

yes but that’s also what makes it regressive

By that same logic, purchasing hamburgers is regressive.

No, you don’t have to purchase a hamburger to earn a living, but you probably have to travel.

You have to purchase food.

Yes, and taxing food would be regressive, just as taxing/tolling commuting would be. (It's for this reason that supermarket/unprepared food is often untaxed.)

Groceries are generally taxed differently because people need to eat. It is not a human necessity to travel long distances to work. It seems to me, if you subsidize travel for low earner you're subsidizing companies that pay their employees poorly, not benefiting the employees themselves. Maybe a subsidy for low earning jobs that are a public good make sense (teachers, for instance), but why should we give subsidies to private for-profit companies to allow them to avoid covering the externalities of their business?

Where their employees choose to live is not an externality caused by the business IMO.

Choosing to open a business that pays minimum wage in an area without housing that is affordable for minimum wage workers inherently generates traffic, doesn't it? I don't have a very solid opinion on all this, honestly, but it seems logical. What do you think?

I suppose that choosing to open a business not in someone’s house generates traffic.

The alternative is not Walmart paying stickers $50K/yr. The alternative is no or fewer Walmart jobs in locales with these laws. IMO, that makes things worse, not better, for prospective Walmart employees.

I also feel like where I live is none of my employer’s business, provided I can do the job. Maybe I’d make an exception for fire or police, but even there I’d rather not make the employer totally paternalistic.

People feel that incentive very differently. If the 101 is a $5 toll road during rush hour it barely changes my habits. In fact, it might actually encourage me to go in earlier because I expect traffic to be better. The cost of time spent in traffic already outweighs any congestion charge - for me.

There are, however, a lot of people for whom that $50 more in weekly commute costs, would be financially deadly and almost unavoidable.

It's easy for me to advocate for congestion charges; they clearly benefit me. Those who make a lot less than I do might feel very differently.

Well, maybe there is a lesson in that. We have to think about what the benefits of each person commuting is. Does it make sense to fill up highways with people going to work in e.g. fast food restaurants?

If the McDonalds employee has to pay $50 in fees to get to their job, it changes their financial calculations around whether that is a viable job. If enough of them decide to quit because of this, the restaurant will need to pay its employees more, subsidize their commutes or shut down.

If the restaurant shuts down, then it didn't make sense for them to be in an expensive urban hub, and the negative externality of the traffic their employees generated is gone. I suspect a lot of only-barely-viable jobs would go this way, and that's where traffic reduction would come from.

If they pay more, or subsidize commutes, they are paying to paying for the negative externality of filling up the highways to bring their employees in. The people who want cheap cheeseburgers in an expensive urban hub will decide if they want to pay more for cheap cheeseburgers to cover the cost. If they don't, it doesn't make sense for it to be there and the traffic isn't worth it.

As a software engineer who shows up at 11, I can tell you that there is a significant current against us late starters. I agree with you that it would be good for us to move this way, but it is hard for all the early birds to swallow.

Ha. "Early bird" is somebody who arrives before 11?

People who schedule meetings before 10am should be fired.

Yes, it is a problem. I can't recall the details, but I have heard compelling arguments why waiting in lines is awful from an economic perspective. You might enjoy hearing an economist's perspective on time.

Poor people would likely benefit overall from more efficient use of roads.

Interstate roads allowed people to live far from things they need (e.g. jobs). Your suggestion is backwards, people should be located near things, not the other way around. Roads with no fees have distorted supply and demand. The market can't function.

Reply on news.ycombinator.com