If that were true, would it not be equally likely that husbands would initiate divorces when their wives lost their jobs, and women would initiate divorces when

If that were true, would it not be equally likely that husbands would initiate divorces when their wives lost their jobs, and women would initiate divorces when their husbands cheat? (I.e., a contradiction of GP's claim - but your comment doesn't sound like you intend to contradict him, but only explain the behavior he described.) You could begin to support such an explanation by pointing out that less women work and so less women are able to lose jobs and so less men divorce them because of it. But that only explains half of the divorce inequalities described by GP. Do men really cheat significantly less than women? So we arrive at a point where either it's more socially acceptable for women to divorce men because of job loss than the reverse (GP correct), or it's more socially acceptable for women to cheat than men (GP incorrect). Conversely, either it's more socially acceptable for men to divorce women for cheating than the reverse (GP correct), or it's more socially acceptable for women to be unemployed than men (GP incorrect). Interesting! Of course, this is all based on anecdote. Also maybe I oopsed and strawmanned inadvertently.