> there aren't regular buses on the side streets because there isn't enough traffic there to justify them
That is the usual problem. But it's usually possible to have freeway buses connect to local buses that drive around an area. Driving around nearly empty buses might seem like a bad idea - but it's the best idea there is when people aren't in a car (because they just jumped off the train or freeway bus). A second solution is having people drive a short distance on local roads from their home to a public transit hub such as a train station, if there is no bus to take them near their homes - The problem is that once you get off a train or freeway bus you might not be at your destination. But that's where public transit can't solve the problem. Zoning and building also needs to change. Without that, people might still have to walk 30 minutes to the office.
The situation where a company puts up an office with a large parking in an "office park" (or whatever it's called) which might be 15 minutes from the closest public transit is the root of the problem. Since people have to use the car to their office anyway, they wouldn't be using the bus even if it existed, as you point out. So the solutuion has to be in both ends: offices and housing needs to be in denser, more walkable areas, for example. I don't think the traffic problems in LA can be solved without fundamentally changing how and where people live and work.
This is my point. There is no amount you can add to the cost of driving that will get people to take a bus to a place the bus doesn't go.
And once you build high density office space next to a train station, people will take the train all on their own without any need to waste resources collecting regressive and privacy-invasive road tolls.
Only if you actually add the buses, and people actually ride them, neither of which is a given.
> Obviously it might be tempting to say "look we can't sacrifice one lane of 4 here to buses when cars are jammed bumper to bumper in the other 3 lanes" but that's how bus lanes work.
That's the problem.
> With too few buses, they are just an ineficcient demonstration to the people in cars that they really should be riding the bus.
That only works if they can ride the bus.
The problem in most of the US is that you have a big congested highway, and then you get off the highway and drive another five or ten minutes on side streets to your destination.
You can't just tell those people to ride the bus, there aren't regular buses on the side streets because there isn't enough traffic there to justify them. And a bus that only takes you to the edge of the congested highway and then leaves you with an hour walk to your destination is useless.
> Whether it costs $0 or $4 to ride the bus to work shouldn't matter
It does matter to the tune of $4, and eliminating fares increases efficiency because you don't have to pay for a fare collection infrastructure, as opposed to bus lanes which dramatically decrease it because you're paying the full cost of a lane and then wasting most of it.
> if parking, gas and tolls are 10 times that for one day of driving
Stop trying to make driving worse. First you have to make the alternatives to driving better, and as soon as you do that, you don't need anything else.
Nobody actually likes driving to work, the only reason anybody does is that the alternatives are broken and terrible.
In order to make driving as bad as the alternatives without making the alternatives better, you would literally end up causing people and companies to move out of the city to avoid the expense. It's the same logic as arguing that burning down the city will reduce traffic congestion. Even if it does, that doesn't make it a good idea.