As I have written before, Phoenix has seen its total transit ridership flat to down since it built its light rail line. This after years of 6-10% a year increases in ridership. Most cities, even the oft-worshipped Portland, have seen the same thing. Here is the chart for Phoenix (if you look closely, you can see how they fudged the bar scaling to make light rail ridership increases look better).
The reason is that per passenger, or per mile, or per route, or whatever way you want to look at it, rail systems are 1-2 orders of magnitude more expensive than buses. Since most cities are reluctant to increase their spending on transit 10-100x when they build trains (and to be fair, proponents of rail projects frequently make this worse by fibbing about future costs and revenue expectations), what happens is that bus routes are cut to fund rail lines. But since buses are so much cheaper, 10 units of bus capacity, or more, must be cut for each one unit of rail capacity.
The Anti-planner shows us an example in Honolulu. No, the line is not finished so this effect has not happened yet, but you can see it from a mile away:
The city and state officials who promoted construction of Honolulu’s rail transit line now admitthat they don’t know how they are going to pay for the cost of operating that line. Between 2019, when the first part of the line is expected to open for business, and 2031, those costs are expected to be $1.7 billion, or about $140 million per year. In 2011, the annual operating cost was estimated to be $126 million a year.
Honolulu has about a hundred bus routes, which cost about $183 million to operate in 2013, or less than $2 million per route. The rail line will therefore cost about 70 times as much to operate as the average bus route.
So they have budgeted no money for operations, and are probably underestimating net operating costs as their revenue projections, as discussed later in the article, are transparently over-optimistic (this is always a good bet, since 99% of rail projects under-estimate their costs and over-estimate their ridership). The rail line will cost as much to operate as 2/3 of their city's entire bus system, which is extensive and well-used. So how many bus routes will be cut to fund this one route? 10? 30? 70?
By the way, beyond the obvious harm to taxpayers, the other people hurt by this are the poor who are disporportionately bus users. Rail systems almost always go from middle/upper class suburbs to business districts and seldom mirror the transit patterns of the poor. Middle class folks who wouldn't be caught dead on a bus love the trains, but these same folks already have transportation alternatives. The bus lines that get cut to fund the trains almost always serve much lower income folks with fewer alternatives.