This post and this post claim the seeming increase in 1980 isn't what it's made out to be - coming from cohort effects, the reversal of a slowdown in 1950 caused by wars/economic depression.

My sense is this is true, and obesity is best thought of as 'having increased since a bit after the industrial revolution', and the increase in 1980 just continues trends like 'food abundance increases, food is desgined to taste better' rather than being a step increase in rate. There is also probably a 'noise + increasing rate looks like a step' effect - if you take a smooth exponential and multiply it by smooth noise, it'll look like there are places when the rate sharply increases because the noise will temporarily line up with the overall trend of growth.

Note that there really isn't a single correct answer here - for reasons like https://slatestarcodex.com/2019/03/13/does-reality-drive-straight-lines-on-graphs-or-do-straight-lines-on-graphs-drive-reality/, which is why this is a bounty on interesting information as opposed to a yes/no market

100-200 bounty per good link / good relevant argument that I find interesting. 1k bounty if my position significantly changes (either towards 'more confident nothing particularly happened in 1980' or 'think its more plausible something happened in 1980'

the "bounty" is 5 because if I put in a 1k bounty that locks up the 1k forever, and this might just not get responses