Using hydrogen to fuel the first stage required building a massively heavy cryogenic tank, impairing performance so greatly the RS-25s cant lift the SLS without SRBs.
And not only will it cost $20B before it’s first launch, but each flight will destroy over a half billion in engines alone. So the minimum cost per flight is over $1B without counting the R&D, about $3B to $5B per flight in total depending on how many flights are flown to amortize the R&D.
And with all that, SLS has very limited capabilities. The block 1 will be the largest heavy lift launch system (until Starship launches), but it’s only about 30% more payload mass than a $150M falcon heavy expendable.
But worse is the maximum possible cadence for the SLS is twice yearly, making it both useless and unnecessary for more ambitious missions that will rely on in orbit refueling. It’s just an archaic dinosaur built out of nearly 50 year old parts.
Let’s say you’re a member of congress accustomed to how programs work and someone says to you, pay me to make a rocket that is totally reusable to make the moon so cheap that tons of people will do it.
For one, cheap? Full stop. In whose interest is it to make space exploration cheap? Does the government want access to space cheap enough that someone might do something destabilizing or disruptive in any way?
Ok, let’s skip past cheap and the Congress-critter looks at the shiny Starship and asks, “How much stuff can it take to the moon?” The answer is “lots if we figure out orbital rocket refueling, otherwise nothing.” To get into TLI and beyond, SpaceX or somebody needs to figure out how to do that. Sure there have been some experiments, but nothing concrete yet.
Ok, you skip past the sci-fi orbital refueling and the suited person says, “I love it and I’ll vote for it, but tell me where all is it going to be built so we now which senators will speed this along.” You say, well, design and manufacturing in CA, and a KSC processing facility, engine tests in Central TX countryside and a new space-port in Boca Chica. Response is “oh come on man—CA, FL and TX already have tons of stuff going on and their reps won’t care about something this small! We need Idaho, Missouri, Mississippi, the kinds of places where a rep will burn down their grandma’s house for a project this size. Come back when you have a real plan.”
While Starship (and maybe New Glenn) does look more attractive than SLS now, a effort as large as Artemis is more concerned with capability than immediately driving the cost down. SLS is a capable vehicle moved forward by proven methods by experienced vendors. The lower risk drives higher and more predictable up front costs.
Check out Everyday Astronaut for some interesting perspective on what is going on with SLS and Starship:
https://everydayastronaut.com/sls-vs-starship/