The question I haven't heard anybody ask about today's big NYC explosion, yet

"But at some point doesn't the system become so archaic and dangerous that it's worth subsidizing a mass switchover?"

short answer... no. Mostly because the buildings that use steam heat don't have boiler rooms --- and the cost of converting to other forms of heating is prohibitively expensive for large buildings (I doubt if they'd allow baseboard heating systems in older, non-"fire proof" buildings). (We have the same kind of system here in Philly, btw).

Switching these systems over is ridiculously easy to do. The problem is that no one wants to pay for it.

So that raises the question: how often does a street have to explode before cost-benefit analysis suggests the system should be upgraded? What is the social cost of an injured New Yorker?

"So that raises the question: how often does a street have to explode before cost-benefit analysis suggests the system should be upgraded? What is the social cost of an injured New Yorker?
"

well, the problem is that the cost of switching over would have to be borne by "private" capital... that cuts into profits, and as we all know private capital is perfectly happen to have people poisoned, injured, or killed as long as they don't incur any costs.

An important consideration is how the steam is generated. Often, there are industrial processes that have heat that would be wasted except that it can be used to create steam for heating.

An example (which may not apply here) is power generation. To spin a turbine, you need high pressure steam. After the team has expanded through the turbine (spinning it to generate electricity), it is no longer high enough pressure to do any more work with. You can either vent it to atmosphere and waste all of the heat that is latent in the steam, or you can put it in a pipe, run it down the street, and heat the Empire State Building.

Even if the steam is generated exclusively for heating, you still can have cost benefits to generating all the steam in one place. For example, one giant industrial boiler is probably cheaper to build and maintain, more efficient, and more reliable than thousands of individual boilers in tousands of separate buildings.

Is steam all that dangerous? For example, is it more dangerous than commuter travel by car? How does NY per capita deaths from steam pipe explosions compare to the death rate of children riding bikes without a helmet? Which would be cheaper to fix?

Money >>> Human life

Actually, co-generation plants (generate electrical power and steam for heating) are quite common. Many co-gen plants are at the front of "green" energy since they can offer a centralized electrical and heating system for a community. This steam would normally be sent into the atmosphere, thus it can be very efficient compared to individual heating solutions - and cheaper. The cost of converting would be very high in existing buildings.

Asking why the system isn't replaced yet when a pipe that is close to a century old blows is like asking why we don't shut down a power plant when a wind storm knocks down a utility pole. Both are dangerous and can be spectacular, but with proper maintenance and upkeep the risk can be minimized.

Do some more research and don't overreact like everyone else.

Umm, Bill, my post was stated as a question. I wanted answers, and I got them. Call it a cheap form of research.