Imagine if you have an IT company with a server. Your server can handle 1000 concurrent users. It is running slowly because a lot of people try to access it.
So you buy another server. Now you have 2000 users, but once again it is starting to run slowly.
You decide to stop there. Clearly, more servers wont help. You'd just get more and more users.
You have 1000 users and a server that can easily do the work 20000 users want it to do each day. Yet, every weekday between 08:00 and 9:00, your server is overloaded.
You buy a second server. Now, your server still is overloaded between 08:00 and 9:00 but fewer people use it between 07:00 and 08:00 and 09:00 and 10:00.
There are plenty of people avoiding the main traffic times for whom the annoyance of congestion is just a tiny bit larger than the annoyance of leaving home an hour earlier. Take away a bit of congestion and they move their schedule.
Cities have relatively fixed populations. The problem is not allowing everyone to drive their own car: that's stupid. The problem is how to persuade more people to catch the bus/train. Building roads doesn't make anyone's life any easier or better because it just persuades people to drive instead of catching public transport. Less roads actually helps because public transport gets better.
Your analogy is broken because building more servers doesn't "get more users". You have to service a fixed number of users in the most efficient manner possible. Providing a car lane for each of them is not efficient.
We all know that roads aren't free. If they weren't worth building in terms of increased economic output, however, then city planners would stop building them.
That seems reasonable on its face. The reality is that state and federal funding hinges on city planners padding budgets and increasing capacity to meet future growth. Meanwhile, cities and towns put themselves in hock to meet the servicing demands of their infrastructure, requiring more state and federal funding, requiring new projects with padded budgets and increased capacity.
Notice that nowhere in there was a requirement that said infrastructure pays for itself. More often than not, and especially in small towns, it doesn't.
You say that as though you think there isn't a significant economic benefit to having a more efficient transportation network. Having higher-capacity roads is beneficial even if they aren't toll roads.
You say that as though you think there isn't a significant economic benefit to having a more efficient transportation network
You're assuming that having more roads makes it more efficient, but that's not necessarily true. If having less urban roads convinces more people to use mass transit, for example, it might be more efficient.
> There was an article about people's tendency to soak up additional road space posted on HN fairly recently
Article provides very little useful information. Again, it's obvious that if you expand one road in a congested city then it will get full of cars again, because it takes the overflow from all of the other congested roads. And if you reduce congestion in general then people might choose to drive more often.
But this does not persist as a 1:1 relationship. China has proved as much by building a bunch of empty highways and cities, as is also demonstrated by urban highways carrying more traffic than rural highways with the same number of lanes. The "problem" is that if you have a lot of congested highways then you need to add a lot of lanes before you'll hit saturation, but that hardly proves that it can't be done. The observed 1:1 relationship in some locations is easily explained as a result of widespread insufficient infrastructure investment, resulting in roads with insufficient capacity even after the expansion.
I had a recording of Ben Elton's standup in the mid-1980s in the UK, and he raised the issue of adding lanes to clogged roads just inviting more traffic, leaving the road just as clogged as before.
He was drawing an analogy to the 'swing-top' bin - how you always try to fit more in and never empty it. "What would be the state of affairs a week from now if I gave you all a brand new swing-top bin to put next to your current one?"
(You can also see something similar with internet bandwidth. Same principle, I think.)
There's a Wikipedia article about the theory that narrower, signage-free roads are safer on aggregate: http://en.wikipedia.org/wiki/Shared_space
(I don't know/care if these are useful to you - simply a brain dump, as you asked)