Spot-on analysis. I agree that transport should operate on a basically break-even basis, but offset in two ways:

1. Where the Government wants to subsidize some group (e.g. help the disadvantaged by giving them discounts) they should pay the fair price to the transit agency out of the budget of Welfare, not drag on the financials of the transport agency. In other words, it shouldn't be possible that the transport agency is insolvent only because most of their customers are paying next to nothing. Discussions about whether we should spend a certain sum on subsidizing the poor to ride the bus/train/etc are purely welfare budget discussions.

2. The Government should move additional money into the system when they realize an expansion of transport helps further societal goals: e.g. congestion pricing funds should help to expand transit, or the government pays part of the cost to build new rail service to reduce congestion on the roads.

> In an alternate universe, public transport is run to compete with the car, and attracts all demographics. Day-to-day operations are un-subsidised, and therefore relatively expensive. It competes on value. People use it because it's a better experience than driving.

The problem with this in the US is that it's nearly impossible for the bus to be faster than a car without making the car slower on purpose, and the latter is the thing which is going to create the most opposition, because you're essentially screwing people over during the transition period -- which would take years if not decades.

In the meantime people still can't take the bus because the higher density housing that makes mass transit viable where they live hasn't been built yet etc., and as long as they're stuck in a car they're going to fight you hard if you try to make being stuck in a car even worse.

Meanwhile, cars are expensive. ~$500/mo for a typical car payment, another $100+ for insurance, another $100+ for gas, you're already at $8400+/year per vehicle before adding repairs and maintenance etc. For a two-car household that's more than 20% of the median household income. Make mass transit completely free and people start preferring the housing where mass transit is viable, which means more of it gets built, which is the thing you need to actually make it work.

It's a chicken and egg problem. The way to make buses competitive is to build bus only lanes. But to do that you end up removing a lane for drivers and dedicating enforcement resources to keeping bus lanes free of private vehicle traffic.

The usual pattern is when a bus only lane is proposed, drivers complain because they view the bus as a social program. Local legislators often take the drivers' side because they also view the bus as a social program. Even if you get the political capital to push a bus only lane, traffic enforcement will routinely ignore bus lane violations. LA is making waves on the latter problem by attaching cameras to buses which automatically write tickets for cars blocking the bus lane.

Ultimately it's a politics problem. If nobody wants to spend political capital on running a bus system as a transport program, it ends up as a social program.

This idea occurred to me while I was traveling in Europe. Many of their trains have two classes of cars, where the first class is just slightly nicer. This could be done with buses too. Just alternate buses on the same route, that are expensive and free. The poor can take the free bus, and those who want a more exclusive social experience can pay for the expensive bus.

I can't make any excuses for the social and class implications, but if it got more people on the bus, it might only need to be a temporary measure.

Busses get tiny subsides in the US.

It’s a large percentage of total bus revenue by design, and a significant expense for some local governments. But the number only look large because of how we split the vast majority of government spending into federal and state budgets with local budgets being relatively anemic by comparison.

Private car ownership is a better everyday solution for almost anyone who can afford it, which includes the vast majority of Americans. If buses tried to compete with cars, they would lose. The only remaining niche for the bus is as a public accommodation for the poor, disabled, and elderly, or occasionally in dense city centers.

At least that’s what I think. But if you’re right, and there’s a version of bus transport that’s viable without subsidy, then there should be a market opportunity for a private business to provide that type of bus transport. This actually exists for long range intercity buses already, but you’d think it should be possible inside of some cities. I haven’t looked into this in a lot of detail but I wouldn’t be surprised if it was effectively impossible to try and start a private bus service in most cities, specifically because that would reduce ridership of city transit and threaten all of the unionized public sector jobs in that system. In which case the bus system isn’t really even for the poor and elderly anymore; it’s for the transit workers union, which undoubtedly is a player in city politics.