A good chunk of the costs of the space shuttle was because it didn't launch as often as it was originally supposed to, which made it harder to amortize R&D cost

A good chunk of the costs of the space shuttle was because it didn't launch as often as it was originally supposed to, which made it harder to amortize R&D costs. The rockets themselves were expensive, and refurbishing the "reusable" parts wasn't cheap. I think SpaceX is in a good position to dramatically lower the costs, as long as they can keep refurbishment costs down and share as much architecture as possible between their manned and unmanned launches. Being able to use a rocket from a manned F9 launch another 20 times (even if they're only certified to use it on unmanned launches) would do wonders for reducing costs.