Estrogen, cortisol and testosterone all operate in a delicate, interrelated balance, so it's not too big a leap.
Take a simple model with a combination of all the factors the author discusses (type of diet, sugar, trans-fats, caloric intake, processed foods, exercise, genetics etc), allow interaction between the factors, and have the lipostat as the dependent variable. I.e. several factors together will, over time, upregulate the lipostat.
A relatively simple model that evades the authors objections, yet might fit the data.
I've sometimes wondered if microbiome factors are in play, and by extension, things like antibiotic use (maybe there are some studies about antibiotic exposure and weight gain?). I've usually thought of it in terms of microbiome -> gut -> digestion though, and the lipostat section made me wonder if it's more like microbiome -> "immune response" -> caloric expenditure.
This study was being discussed quite a bit when it came out, although there was prior work on it:
https://elifesciences.org/articles/49555
https://www.scientificamerican.com/article/are-human-body-te...
One hypothesis is that there's been a decline in baseline immune activity with decreased exposure to pathogens, better sanitation, better treatment of infections, etc. But people don't know for sure.
I'm not sure that it applies to the obesity crisis as temps have been declining over a longer time period than what seems to be involved in that. But it got me wondering if there's something like that that might have came into play later. My guess is no, that it's something else, but I could see how antibiotic use might somehow be involved.