Hacker News Reader

Comment by yadaeno | Hacker News Reader

And its one of the best investments the government ever made. 600% ROI transforming the American economy, connecting communities.

The amount of displacement caused by a national rail system or bike lanes would be minuscule in comparison but have a similar impact. Politics nowadays seems too risk averse to pull large infra projects. Sometimes you have demolish some homes but nobody is brave enough to make these tough decisions anymore.

https://www.google.com/amp/s/infrastructurereportcard.org/ha...

View on HN

It's possible there are far more negative externalities than you're accounting for. What's 600% ROI if it turns out it contributed significantly to inequality, both for income and race? How negatively has that effected the nation, or even just the economy, over decades?

It's important to note that the argument is not as simple as "highways or no highways", but more nuanced, as perhaps 10% harder to accomplish yields a 50% reduction in problems. Finding that sane middle ground is hard, but we should be careful not to reduce the problem to such simplicity that the many of the important parts of the solution are no longer even assessed.

>>What's 600% ROI if it turns out it contributed significantly to inequality, both for income and race?

I don't see improvements in the physical economy appreciably increasing inequality.

Most inequality seems to step from housing scarcity driving up housing costs. [1] (see Figure 3)

Improvements in transportation could even mitigate this sort of inequality by overcoming spatial misallocations caused by housing scarcity in the highest productivity zones. [2]

[1] https://www.brookings.edu/wp-content/uploads/2016/07/2015a_r...

[2] https://www.aeaweb.org/articles?id=10.1257/mac.20170388

The context you're missing, which is being alluded to in many comments, is that when the highways were put in place and they chose to put them through cities instead of around them politicians and planners had to choose where they went and what neighborhoods they bisected and displaced. Which neighborhoods do you think bore the brunt of that problem? Where do you think they decided to put the highways that made it harder to reach different parts of the city, by relegating city traffic to specific overpasses and underpasses?

Class and race had a lot to do with it in many cases. You've possibly heard of "the wrong side of the tracks". That statement could be considered even more accurate in many modern context if stated as "the wrong side of the highway". The interesting part is why it's apt though, which is what we're talking about.

For one take on this, see https://www.history.com/news/interstate-highway-system-infra...

> highways that made it harder to reach different parts of the city

One blatant example of exclusionary infrastructure was Robert Moses building bridges in New York that were too low for busses to pass under, in order to discourage travel to his parks and beaches.

I can also recommend the Well There's Your Problem podcast: https://www.youtube.com/watch?v=8oq0u2i4iHc

Highways certainly boost ROI, but there's devil in the details.

The interstates that linked up major cities certainly boosted investment. But Europe also built a network of such highways; the major difference is that they built their system to go mostly around cities, whereas the US just plowed them straight through cities. (The reasoning being that you don't really want to subject traffic passing through to high levels of commuter congestion, you want to minimize impacts to high concentrations of people and businesses, land acquisition is cheaper outside cities than inside, etc.) Because they did not impact nearly as many people to the same degree, Europe didn't experience the same overcorrection in terms of public process. (There is NIMBY-ism in Europe, to be sure, but they are definitely more successful at actually building things and keeping costs low compared to the US.)

There is also the subject of diminishing returns. Cutting travel time between the coasts from weeks to days is certainly a huge return on investment. Today most expansion projects look like "let's cut 30 seconds of waiting at this traffic light" at a cost of hundreds of millions of dollars at times. The report you linked is from ASCE, which is not exactly a neutral party for infrastructure development. Here's a critique of an earlier report from the same organization:

> Consider the following from the report:

> ASCE estimated the "costs to households and businesses" from transportation deficiencies in 2010 to be $130 billion. (page 3)

> ASCE estimated the cumulative losses to businesses will be $430 billion by 2020. (page 5)

> ASCE estimated the cumulative losses to households will be $482 billion by 2020. (page 5)

> If you add these together, the total cost to households and businesses is $1.042 trillion. Well, ASCE states that to reach "minimum tolerable conditions" (a pretty sad standard) would take an investment of $220 billion annually. Over 10 years, that's $2.2 trillion. Yeah, you read that right. The American Society of Civil Engineers wrote a report suggesting that over the next decade we spend $2.2 trillion so we can save $1.0 trillion.

It's kind of hilarious that that's your takeaway from this. Where I live in Dallas, all of downtown is walled off by elevated and sunken highways with only one or two bridges across them, all but guaranteeing adjacent neighborhoods are completely cut off for pedestrians, to prevent them from ever becoming larger communities. There was no reason for this whatsoever except the east, south, and west side neighborhoods were all predominantly black and planners wanted to keep them out.

We're still suffering from these decisions today, with all manner of proposal to make the area walkable dead in the water due to the expense that would be entailed by rerouting or rebuilding highways.

There is no return on investment anyone got from this that couldn't have been gotten from highways moved another mile or two away into the purely industrial districts where they wouldn't have disrupted anybody.

> And its one of the best investments the government ever made.

I don't think the experiment is over.

The benefits of a highway-based society have been reaped during a period of 1) cheap fossil fuel energy and 2) cheap land. Both of these are becoming more scarce which threatens to tip the ROI negative. Strongtowns.org has done some research that most American cities and suburbs will be facing insolvency in coming years because their revenue is insufficient to cover the costs of a road-based infrastructure (including water and power lines) that scales quadratically.

To say that highways have been a good investment you would need to compare them with alternatives and examine their opportunity cost. A major problem with car transport is that it has worse scaling than most public transport: each household must personally invest in a car, garage, and fuel; private businesses must invest in parking, which drives up real estate prices; each car decreases the throughput of the roadway, requiring more roads to handle more cars--and when you run out of space for roadway then commute times increase. With most aspects of car transport you never get better than linear scaling.

So the real question is: if you took all the money that has been spent on cars, roadways, fuel, repairs, etc. would it have been better spent on trains, buses, or some hybrid system?

The Interstates between cities are a modern marvel whose impact can barely be overstated; the Interstates within cities were an unnecessary, destructive mistake.

Are interstates just supposed to end once you reach a city, and in order to continue on you’re supposed to navigate a series of roads that take you to the other edge of town, where the interstate starts again? Seems less than ideal.

Going around is an option.

We could have built the interstates around urban cores and connected them to existing thoroughfares within cities. Most of our urban cores have expanded out to where those highways would have been built then, anyway. There was no need to go right through existing neighborhoods.

Every city in the path of an Interstate should have a ring road around it, instead of cutting the city in two or more parts. Basically, a giant roundabout.

This is traditionally accomplished with ring roads that go around the city. Take Minneapolis for example, you'd be crazy to take I-94 through the heart of the city. If your driving through on your way out west your going to take the 694 bypass.

It is, but less so on the East Coast. Not enough to make the point irrelevant, but enough to complicate the generality some. Like, if you're going north from southern New England, for example, you're probably taking I-93 straight through the city. Which is a lot better than it used to be thanks to the Big Dig (which gets so much bad press but really is a great example of an infrastructure project that did, eventually, work), but is still not pleasant, and it's because taking 95 (or, further out, 495) all-the-way-around is as significant detour.

Circumnavigating the cities would be a reasonable alternative, and the Interstate system has incorporated many such "belt roads" since its inception. In retrospect this would have been a much better option than running them through downtowns.

Correct, interstates are not supposed to directly visit every home and business in the country.

The system was engineered to incorporate both loops and spurs, and it's encoded in the numbering system:

I think most folks know that 2-digit interstates with an even first digit are east-west, and with an odd first digit are north-south.

But 3-digit interstates exist too. An even first digit means it's a loop around a city (like I-475 around Flint), and an odd first digit means it's a spur into downtown. (I-375 into Detroit). These numbers can be reused, for instance there's another I-375 in Florida.

I think the argument here is that the _default_ should've been to loop around cities.

According to this article https://en.wikipedia.org/wiki/Interstate_Highway_System#Prim... there is no encoding in separate digits, only the whole number being odd or even has meaning. E.g I-10, I-14, I-30, I-70, I-72 etc are east-west and I-25, I-45,I-65, I-83 etc are north-south.

Read down to the next subsection, "Auxiliary (three-digit) Interstates".

Did not find anything about 2 digits ones there, do you have a particular quote?

Weren’t a lot of those loops absorbed by later growth? Much like siting an airport, it’s hard to pick a route that will always remain outside a city without just going to the middle of nowhere.

Wikipedia says I-280 was planned in 1955, and I think Silicon Valley was mostly still fruit orchards at the time. The silicon transistor would be invented one year later.

The main thing favoring bypasses is that orchards aren‘t neighborhoods full of people, so you avoid the initial destruction and displacement.

What happens after the fact isn‘t as important.

Practically every major city with an Interstate has what is called a bypass, typically at least two. These look like a ring around the city in most cases and are (typically) the even numbered 3 digit interstates.

The bypasses are almost always faster to take if you're just moving on through to the other edge of town. These would have been totally sufficient in almost all cases w/o plowing the Interstate itself through/near the downtown cores in addition to the bypasses.

Every major city that I’ve been to has at least one and sometimes more than three loops around it. You drive down the interstate then take the loop around and back on the interstate to avoid city congestion. It’s like a round-about with the city in the middle.

Have you ever visited an urban area that has been cut up by a highway? I've never seen one that wasn't a dodgy neglected hellhole, even in traditionally denser east coast cities. The affected communities end up disconnected rather than connected to elsewhere, and the outlying areas that do end up being connected could just as well have been connected by a highway routed around the urban area.

There's an argument to be made for cut and cover construction and then putting in a park or newer homes or something, but surface highways through cities are not a goal to strive for.

Parent’s comment stayed neutral, but the homes and area that were destroyed weren’t from some random population. In almost every cases it’s poorer and/or minority neighborhoods that were run over and/or separated from the rest of the town by a huge highway, with redlining making sure they don’t come back the other side.

“connecting communities” only applies to some specific flavor of communities, and the ROI mostly benefited these specific groups as well.

It's brave now to deliberately choose to route infrastructure through neighborhoods that have minorities living in them?

I think with today's politicians it's more of an issue of credit. These projects take a very long time, piss off people until they're done, and there's a high chance a different politician will be in the office at that time and claim credit.

Reply on news.ycombinator.com