I read the whole thing.
I'm baffled that statisticians would come up with this.
I can't cut and paste the relevant bits as it puts a load of blurb.
The relevant section is 'Proposed solution – ‘gender by default’ principle' near the beginning.
Apparently in policy making etc, gender is more relevant than sex. Also they note that lots of people think gender means sex. So they are banking on people putting their sex to this question.
They say sex and gender can be asked if relevant, but this is an exception.
The whole thing makes no sense.
And is wholly focused on capturing stats about trans people. Women aren't mentioned at all from what I can see.
I really don't understand how this totally illogical, disconnected from reality, situation has come about.