Predictive analytics may make insurance less socially fair. Seems great to know who is likely to default on a loan/get sick/crash their car. But if you knew perfectly, then it wouldn't be insurance at all -- no redistribution of risk. Partial knowledge means less redistribution.
-
-
-
Replying to @zeynep
Here's the only work I know that analyzes the effects of better payback prediction on mortgage availability and interest rates. Basically, it doesn't really change the averages but increases the dispersion of the outcome. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3072038 …
1 reply 1 retweet 3 likes -
Replying to @jonathanstray
That’s interesting, thanks. This is (yet another) a major issue we haven’t grappled with. There is apparently a forthcoming book on the history of some of this, by Caley Horan. https://www.wired.com/story/ideas-joi-ito-insurance-algorithms/ …
1 reply 0 retweets 0 likes -
Replying to @zeynep
Fascinating, thanks. OTOH that paper discusses evidence that online lending has reduced discrimination -- and unjustified differences in availability and costs -- by reducing the scope of human loan officers.
1 reply 0 retweets 1 like
I can see both operating at the same time.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.