C2, my purely instinctual guess is that the numbers we have out of China are not correct. Whether that's on purpose or a flaw in reporting or whether it took them a while to figure it out - that is anyone's guess.
I also wonder why some states in the US (like NC and NM) have a .02 death to positive test result ratio and others, like NY and WA have a .05. There must be either some odd health belt or poor reporting or (most likely) the cases started later and they locked down when the rest of us did, even though they hadn't had serious outbreaks.
In my opinion it's also partly that the SARS Cov2 virus was only sequenced in mid January (like the 20th, maybe?) and that there were likely already deaths and recoveries some time before that in some areas like NY and WA, but they were attributed to "flu like illness" and pneumonia - before we had testing.