> The study compares energy expenditure across different economic groups i.e. western people sitting in offices versus hunter-gatherers in Africa, and found that difference in energy expenditure does not account for differences in obesity, so points to consumption as the likely reason.

That seems like a kind of large assumption to make. Obviously it seems like it has to be either diet or exercise, but if the obvious answer was always right we wouldn't need to do studies in the first place.