[Alon]

The driver is a big part of the cost of the bus, but not so overwhelming that railstitution can save that much money just by using larger vehicles. I don't know the numbers in Vancouver, but in New York, the operating costs of the NYCT bus system are $2.3 billion a year, there are 11,000 bus drivers, and the annual pay per bus driver averages a hair less than $60,000. So operator salaries are a bit more than a quarter of operating costs. It's sizable, but it means larger vehicles aren't a magic bullet. On top of that, larger vehicles only save you money when you're reducing frequency, which is possible at rush hour since the bus route in question comes every 3 minutes but isn't possible on weekends or in the evenings when it comes every 10 minutes without cutting ridership.

The reasons to railstitute are not exactly saving money on bus operators. Rather, they're,

1. Smoother ride.
2. Higher capacity with larger vehicles.
3. Lower operating costs on maintenance, energy, etc.
4. Higher speed as long as there are dedicated lanes (if there aren't, buses can be faster by swerving around obstacles).

Operating costs go down when speed goes up because you have fewer operating hours for the same service, and usually also more passengers to spread the costs among, but mixed-traffic streetcars don't have that advantage. Calgary's light rail has very low operating costs per passenger, but that's by keeping speeds high outside the downtown core by using separate rights of way. A streetcar that averages 20 km/h won't have that advantage.

Broadway is a very high-ridership corridor, but it's uniquely sucky for streetcars essentially because it's too high-ridership. It's a bit like Flatbush, Utica, and Nostrand in Brooklyn: a streetcar even with dedicated lanes won't be able to absorb peak demand, requiring to keep rush-hour relief buses, eliminating any operating cost advantage of light rail over buses. Vancouver is also a uniquely sucky city for light rail because, for reasons I don't entirely understand, the subway:LRT cost ratio is very low, about 2 whereas in most cities it's 4-6. (Vancouver at least can get away with one subway; Brooklyn would ideally need two, one on Nostrand and one on Utica.) The issue then is that the low-cost at-grade option in Vancouver isn't that low-cost and doesn't solve that many problems.