SpaceX Dragon successfully docks with International Space Station

According to NASA[1], a shuttle launch cost $450 million on average. Their contract with SpaceX covers 12 launches for $1.6 billion[2], or roughly $133 million per launch. SpaceX also received $75 million from NASA to develop a crewed version of the Dragon capsule.

I'm sure NASA has made other payments to SpaceX over the years, but even if those payments totaled a couple billion it's still cheaper than 12 Shuttle launches just to supply the ISS.

According to Wikipedia[3], the Falcon 9 rocket itself costs half as much as an Atlas V.

[1] http://www.nasa.gov/centers/kennedy/about/information/shuttl...

[2] http://www.space.com/13012-nasa-unveils-private-space-taxi-c...

[3] http://en.wikipedia.org/wiki/Comparison_of_orbital_launch_sy...

The easiest direct comparison would be to the ESA's ATV. Each ATV launch costs around $900 million in amortized development costs plus vehicle costs plus launch costs. But, it delivers 6,595 kg of cargo, which puts the delivered cost of cargo at about $132/gram.

In comparison, the Dragon is on a fixed price contract (about $133 million per vehicle) which includes launch and operations and whatnot. This particular Dragon is delivering 1050 kg of supplies (677 kg within the pressurized part of the spacecraft, the rest within the unpressurized "trunk"). This comes out to a cost to NASA of $127/gram.

Which, admittedly, is not Earth shattering. However, this flight will also be returning 1370 kg of equipment back to Earth, including highly valuable scientific experiments and such-like. And Dragon is currently the only vehicle capable of doing so. On the whole this makes Dragon more than worthwhile.

However, this is still only the 2nd operational Dragon resupply mission, and they will continue to increase the amount of cargo delivered as they become more confident in the vehicle and especially as the Falcon 9 v1.1 becomes operational (which will happen in the next flight) which will add about 4 tonnes of additional payload capacity. If fully utilized it could bring the cost of cargo deliveries to ISS down to around $30-$50/g. Also, there is definitely an element of subsidizing the development of new systems in the CRS contract terms, though even so they're reasonably cost effective.

Edit: to compare to the Shuttle, on an MPLM mission the Shuttle could deliver about 15 tonnes of supplies and equipment and cost around $1.5 to $2 billion per launch (the higher figure due to the extremely low flight rate near the end of the Shuttle program), which works out to a cargo delivery cost of around $100/g, though the Shuttle could also return significant amounts of cargo.

Edit2: I inadvertently omitted the cost of the MPLM modules themselves from the calculation above, although including them would be difficult since most of them were built by the Italian Space Agency as payment in kind for ISS access, overall they would only affect the cost figures by less than 20%. Also, fun fact: the MPLMs were named for famous Renaissance artists, or ninja turtles, here's NASA's logo: http://en.wikipedia.org/wiki/File:Multi-Purpose_Logistics_Mo...