6% or ~500 lives per year according to this article (though it notes an increase in vehicle related deaths, so some of that may be cancelled by the additional road traffic) - https://qz.com/2038153/drunk-driving-deaths-dropped-as-more-...

Not a bad result, but probably not worth the problems that their current business model entails. There's clearly opportunities for better transportation options, they just need to make economic sense.

There's a lot of research and much of it is conflicting, here's two examples with vastly different conclusions

https://jamanetwork.com/journals/jamasurgery/fullarticle/278...

https://www.nber.org/papers/w29071

Both researches show benefits but the nber one somehow twists it into a different narrative. I understand why hackernews users hate uber, but shouldn’t professional economists try to be unbiased?

The point of the nber study is that there's a ton of varying answers out there. If it's biased towards anything it's "good science."