3 ms·
That is a statement that really needs citation and qualification to back it up. I'd argue that 'cheaper' is used in a vacuum here. By that I mean that the point
by jmward01 2y ago
That is a statement that really needs citation and qualification to back it up. I'd argue that 'cheaper' is used in a vacuum here. By that I mean that the point of mass transit isn't how much it costs, but how much value it provides. By that measure the bus services that replaced streetcars and other mass transit really doesn't stack up. Busses have led to much lower ridership which has led to a massive amount of bad secondary effects. Looking at how congestion pricing in NYC has increased mass transit use AND economic activity it is pretty clear that 'cheaper' has led to secondary effects that far outweigh any narrow operational gains from switching to busses.
- flomo 2y agoThe citation is every American transit system in the 1950s. Even SF only kept the streetcars where they couldn't replace with a bus line. I think you're making a different argument, where trains attract a more well-heeled commuter. Which is why many cities have brought back LRT as part of a redevelopment plan.
- jmward01 2y agoThe argument I am making is that you can only say something was cheaper if it provided the same level of service or better for less money. In the case of the bus transition it provided worse service as indicated by utilization dropping. I am also making the argument that evaluating the value of transit shouldn't rest entirely on the cost of that service and ridership but on the value as a whole it brings to a city. I mentioned NYC because the evidence there (and in other cities that have implemented congestion pricing) is that as ridership goes up the economic, environment, and social health of a city also goes up. Point being, the bus transition had a very negative value impact. I will also add a final argument, as your ridership drops things like busses may appear less costly per ride simply because you are loosing volume and low volume routes are likely easier to service by bus so, again, cheaper but not an apples to apples comparison.
- flomo 2y agoYes, I like trains and wish we had better transit. I'm responding to the claim streetcars were "torn up for cars", which was not really the case (and frequently subject to a conspiracy theory).
- rsynnott 2y agoSo… it kind of made sense at the time (if you ignored a couple of factors which didn’t become obvious until much later, and which ultimately made it a disaster), and it’s not just an American thing. In 1928, Dublin had 28 tramlines. By 1950, it had zero, with the tram routes being replaced by bus routes (many older Dublin Bus routes basically follow old tram routes even today). It didn’t have trams again til 2004; today it has two lines. But _at the time_, buses were very competitive with trams; there wasn’t all that much other road traffic, so they weren’t really slower (and they could use the tram lines, while they existed, as a right of way), the trams of the time were non-articulated, so buses had similar or higher capacity, and they were more flexible. It’s easy to see how it happened, and it only became apparent that there was a problem when it was far too late to reverse. There are really two problems with buses vs trams; buses are slow when there is lots of other traffic, and, _with modern articulated trams_, higher total capacities are achievable (one of Dublin’s modern tramlines has trams that take over 400 people with a frequency of under 3 minutes between trams at peak, which is a level of capacity that you just can’t really get with buses). Neither of these were factors when the first-generation trams died. Very few tram systems worldwide actually survived from the early 20th century to today; while there are lots of trams today, most are second-generation systems put in from the 80s on.
- seoulsubway 2y ago[dead]