The problem with models is that depending on your assumptions you can show just about anything. In general I think the foundation of a good argument, and good science, is assuming the minimum and seeing if your idea still works. They assume:

- hospital reimbursement drops to medicare rates ($296 billion)

- pharma prices decline 51% ($378 billion)

- administrative overhead way down ($286 billion)

- fraudulent billing way down ($286 billion)

- people stop going to the ER and stop being hospitalized as frequently, so the expanded use of services only costs $198 billion

Those seem like extremely optimistic assumptions, and the paper gives 0 consideration to most knock-on effects. For instance I don't think it's reasonable to cut hospital funding by $296 billion, increase the access to these hospitals, and assume everything will work out just fine. They at least acknowledge not acknowledging this, but that one factor alone already is likely to dramatically reshape things. And there are many others.

(1) we don't know the claims of the study will be accurate if implemented and (2) we would need studies modelling cost savings of alternative approaches to actually compare anything.

Obamacare made wild claims about cost savings as well. That didn't work out for many reasons that could show up again with another universal healthcare push.