As indicated above, I sent Mr. Kasden a request and link to the article. I asked him if he could both confirm the numbers (and give us a source) and confirm whether or not he was accurately quoted in the article.
The key issue IMHO is attributing the CFR numbers to the CDC. They don't appear to be rough guesses (2.1 and 3.3% are fairly precise numbers). If he had been quoted as saying that the CDC estimated 2 to 3%, then I would feel better about it. So, it appears that someone did a calculation and came up with the range, which was then used to calculate the total number of expected deaths.
The important question is, whose numbers are they? Many folks on this site look at this CFR range and consider it realistic, but this would be a huge departure from previous estimates released by the CDC and if they did provide those numbers, it would have to mean they have data that strongly indicates a major shift in the pandemic.
One more thing, if this was a "worst-case" planning scenario, then why the range? It would seem that a range would cover best expected case to worst case.
I'll report back immediately if I hear back from Mr. Kasden.