I think a distinction has to be drawn between people actively disliking a work because they know it was written by the opposite gender vs relating more to a work of fiction written by their gender that emphasises how it feels to live as their gender (Hemmingway, Mishima, Wilde on one side and Austen, the Brontës, and Atwood on the other) and also the high barriers to entry for any woman looking to become a truly respected author that led many of them to pretend to be men. Most men like great works written by women and vice versa, they just feel more comfortable reading things in their comfort zone written by people they relate more to; the lower brow the books are, the more stratified this divide seems to get. The divide also increases the younger a person is

Somewhat interesting side note, the number of men and women reading 'high brow' stuff is roughly the same, but (like the article says) i see a lot less men than women reading low brow pulp stuff (presumably men looking for low brow stuff go elsewhere) so as a percentage of their reading population men seem to buy more fiction written by women than the other way round (but like i said earlier, the women who read high brow stuff read much closer to a 50/50 split than the men who do). The divide also seems to decrease the older the customer is too, before having a rebound when they're getting really ancient, I don't really know the implications of that but it seemed worth mentioning.

They also frame it as biological rather than social. It would clearly be a social phenomenon.