That benefits me a great deal!
Maybe when you said insurance you meant to say "health insurance" but the same principle applies. If I have a heart attack then treating it will be hideously expensive, but that's also unlikely to happen.
"Spreading the risk" doesn't work when the likelihood of something bad happening approaches a percentage in the high double digits, especially with an ageing population and falling birth rates to keep paying into the ponzi scheme.
Historically, instances of entire cities burning were commonplace enough to be noted: the Great Fire of London (1666), the Great Chicago Fire (1871), the Baltimore Fire of 1904, the Great Fire and Earthquake of 1906 in San Francisco.
The Oakland Hills Fire of 1991 was actually the biggest urban fire in the US since the 1906 San Francisco event.
There's a list of fire-related losses here, though it includes terrorist (9/11) and industrial accident events: http://www.nfpa.org/news-and-research/fire-statistics-and-re...
Among the changes I can think of:
1. Better alerting systems.
2. Standardisation. Baltimore and Oakland Hills both saw incompatible hose-hydrant couplings.
3. Building standards. Everything from fireblocks and exits, to sprinkler systems.
4. Replacement of candles, oil lanterns, and gas lighting by electricity. Electrical fires happen, but they're far rarer. The Great Chicago Fire was caused by a lantern.
5. Safety and certification ratings of equipment, appliances, and wiring. "Underwriters Laboratories", the source of the ubiquitous (in the US) "UL" logo, was created by insurance companies.
That is: appropriately collectivising risks can result in reduced overall risk.