It reduces the cost by being fully reusable - getting 200 launches out of each airframe at least. So far rockets:
a) Are only partially reusable in the case of those that can reach orbit.
b) Have a high failure rate.
c) It's not clear that any bits have actually flown twice other than on the suborbital Blue Origin.
d) have more limits about where they can be flown from.
Against that, it's expensive but against that it may have various spinoffs (e.g. military) which will reduce the portion of the development cost that has to be paid back entirely by launcher sales. So there are various ways to look at it.
I think it would be silly not to develop the engine and then assess from there.