3 ms·
Just doing a wee bit of economic risk analysis seems to be beyond this publication. We can absolutely engineer a home to be resilient against <pick your sigma>
by calmbonsai 2y ago
Just doing a wee bit of economic risk analysis seems to be beyond this publication.
We can absolutely engineer a home to be resilient against <pick your sigma> natural disaster short of a tsunami, volcanic eruption, or meteor strike, BUT it makes zero economic sense to do so beyond <pick your sigma>.
The rational approach is to build to <lower sigma> and insure to <higher sigma>.
Many Roman civil engineering structures still exist today and they really shouldn't. They were extremely over-built due to the ignorance of materials science and structural engineering at the time.
Just 'cuz we can doesn't mean we should.
- dustyventure 2y agoThe article is about an area that gets hit with hurricanes multiple times per year and the article ends with a discussion of how poor insurance was for the neighbors. Real estate prices can be very divorced from building costs so I'm not really sure that having terrible standards is giving anything back to the home owners, it might be skimmed as profit for developers or it might mean more money going into a mcluxury race in low quality housing stock.
- calmbonsai 2y agoI think these people simply had bad insurance. I've dealt with multiple home-owners policies over the years including one claim for a hurricane-powered tree branch piercing completely through the roof & window of a bedroom and one case of a complete under-foundation water main break resulting in a "write-off" flood of the basement and parts of the foundation. In both cases, there was an explicit clause related to "re-build as-new to extant code" and, though it took time, everything was re-built as new--including to new code related to water-main piping, electrical grounding, and foundation cement sealants. I concur that having terrible baseline building codes and dwelling underwriting is long-term lose-lose.