I think a way to frame the question is: "Will the coming decline in ad-spend result in proportional decrease in sales (while factoring in crisis driven declines)"
I suspect that this forced experiment will result in identifying that 20% of ad-spenders were getting 80% of the value, while the rest were either not technically capable or just trying to keep up with trends, thus bidding up the ad prices. The result being huge profit for Google/Facebook as the market maker.
[0] https://thecorrespondent.com/100/the-new-dot-com-bubble-is-h...
So that's one way to find out.
It can be observed yet you don't know what is happening.
Big corporations throw money at different purpose and once the budget is out they could be under a directive to just spend.
Or, maybe you aren't subsidizing it and just tossing money in the wind...and that's the problem, you have an x +/-/*/^ y effect and can't separate the two.
So the advertising becomes "locked in" because it was never extricated from the rest of the product and it becomes too much of a risk to pull it.
Some people try to isolate things and figure it out but it's not really possible, externalities change, contexts change, needs change, advertising happens on the sneaker net without your knowledge... It's a real bitch