Even 5B for a novel, and uniquely powerful observation platform would be a bargain!
When you are engineering a unit of 1 pushing the boundaries of science, with multiple conflicting constraints, funded by a variety of self-interested stakeholders, and are forced to do commercial production, rather than govt production, even when it is most cost effective, it isn't like you are heading towards lowest cost, technically acceptable.
It's easy to throw these big numbers around, but do you ever stop to consider the actual cost? Let's do some back-of-the envelope calculations to help us consider the human cost of a 5B telescope.
The Median US household income is ~68,000/year[0]
The Average income tax paid by someone in the 50-75K income range is $4,600/year.[1]
The average working career is probably around 40 years.
5,000,000,000 / 4,600 / 40 = 27,173
To fund a $5B project, 27,173 people (more, actually, since this is household data) could have worked for their entire working lives, with every dime of federal income tax being spent on that one project!
I agree that the JWST is a worthwhile project, but let's not pretend that it's a bargain.
I think that NASA's budget, unlike many other federal agencies, is actually highly beneficial to US Taxpayers due to the large amount of IP/new inventions generated. Similar to DARPA, there are huge ROI factors that come in from an organization that researches, conducts science and engineering efforts on a massive scale, resulting in technologies that consumers can use economically.
Due to the huge amount of technologies generated, and refined from these large prestige missions, I do not consider them to be a waste of funding.
One great example are the weather satellites generated by the NASA/NOAA partnership, such as GOES-R, and JPSS, and their predecessor missions in GOES & POES, to name a few. While they are very expensive they equip meteorologists with the rich data needed to make accurate observations. These observations directly impact human life, both by guiding evacuation decisions, knowing tornado tracks, and also, farming decisions. This same data is used for supply chain management, and there are a number of other uses for it.
Although many commercial media sources will be happy to provide you a weather feed, they often do not tell you that they have a backend connection to NASA, NESDIS and NWS, in order to provide their own weather data, or data from a research satellite. Or they'll provide you a customized photo which is actually a tailored version of imagery from GOES-*.
Because of the incalculable costs of an earth impacting asteroid, or a Carrington-dwarfing electromagnetic storm https://en.wikipedia.org/wiki/Carrington_Event , outward looking to see more of the cosmos is one of the best things we can do to ensure our survival as a species. The more we look out, the more we are able to prepare for such an event.
You can get a pretty good idea of quality of project management based on how well or poorly they estimated project baseline costs at outset.
These are linked - incompetence in estimating costs / complexity = incompetence in execution = insane cost overruns.
And you would get far more science with 5 $2B projects then one project like this. And if this thing has a launch of deployment problem all eggs in one basket. If there are cost overruns and delays, also all eggs in a basket and no other options.
"JWST is now estimated to cost approximately $9.7 billion and launch in October 2021, which represents cost growth of 95 percent and 88 months of schedule delays since the project’s cost and schedule baselines were first established in 2009."
And what the ROI is. Tens (hundreds?) of thousands of dead people and stalling the taliban for a little bit versus unlocking mysteries into the origins of the universe
A cool project, but if you think of the thousands of folks who didn't get funding so this thing could gobble everything up - these projects really become crazy budget wise (SLS did the same path).
I wish they would do pay for performance deals. We'll give you $4B if you put a telescope in space of X size that meets some basic specs.
If you look at commercial side, space imaging (earth facing) has just exploded and the cost side has gotten very very good. So it's clear you can get optics and sensing into space for a lot less.
Given this is standard govt contracting - I'm sure it's come out much higher.