Not the launch cost, the cost of the satellite.

JWST is 6500kg. Quick googling shows SpaceX costs about $3k per kg, so the launch cost is about $20M.

The total project cost for JWST is $10bn according to wikipedia.

Having worked at NASA, let me tell you that number is 100% BS on all levels. You have to go out of your way to make a satellite cost that much, which they did.

Look at how, say, Planetary Resources was putting together their much smaller telescopes and you'd get an idea of how it could be done on the cheap. Then to scale up you'd focus on cost effectiveness instead of maximizing capability at all cost. Do multiple cheap launches ($100m each) instead of spending billions on a complicated automated unfurling mechanism, etc.

I bet SpaceX could assemble a huge telescope in space from smaller telescopes, and they could do it for much less than NASA ever could. NASA is a jobs program.

> NASA is a jobs program

NASA is first and foremost our nation's science program. Everything from the network card driver your AWS instance is using to the planes flying into hurricanes to determine how strong they will hit was built with major contributions from NASA engineers.

LASIK eye surgery, the nutritional supplements in baby formula, solar panels affordable for consumers, OpenStack, even the BowFlex were all developed by NASA or under NASA funding programs.

You could at best say NASA is a jobs program for HN readers. :)

It is correct to call NASA’s human space flight a jobs program. It is specifically engineered to maximize the number of jobs spread across key congressional districts and states. And unfortunately this makes it extremely wasteful.

It only costs that much because launches are expensive. If you live in a world where your satellite costs $300m+ to launch your going to spend a lot on the satellite to make sure that it is reliable and has a long life.

If suddenly launches only cost $50m then you might build a cheaper and slightly less reliable satellite as you know you can always launch another for only $50m

With cheaper launches, you can throw up several simpler instruments per year, instead of a FANTASTIC one per generation.

Or with cheaper launches you can assemble it in space instead of using that crazy elaborate unfolding mechanism that accounts for much of the cost.