The cost of developing 3 aircrafts is less than linear with the cost of developing one (there are some commonalities). The cost of developing one aircraft that meets the requirements of all, that can be astronomical, assuming a solution to that problem exists. Let's say you want a motorcycle-truck-helicopter that rides on rough terrain like a cross bike, carries 40 tons of goods like a truck and flies. Good luck with that :D

Designing a plane that can be used by the Air Force and Navy can be done; adding the vertical flight to it, you break it: it makes it fat (aerodynamic), heavy and all this for a use case that is rare and useless to the other branches. For example, an F18 can be used by USAF and Navy, a modified F16 can be adapted to carrier operation with a reasonable cost and still you have good planes. Take one of these and put the VTOL in it: there is no high-performance VTOL plane ever built, so what are we dreaming about?

I don't have raw numbers, but last I checked the cost estimates of maintaining these aircraft over the long term is still projected to be cheaper than the alternative.

Some of the aircraft flown by the USAF and the US Navy have higher annual operational costs than the sticker price of those same aircraft. The F-35 is an attempt to avoid repeating the same mistakes, but obviously, easier said than done.

Why build three good planes when you can have one shitty one for four times the price?