Obviously there are abusive cases where pharma companies are spending marketing dollars to promote products that aren't even marginally better than the alternatives. But it seems to me like there are broad cases in the industry where --- whatever other criticism you might want to level about marketing spending --- marketing simply isn't ever zero-sum.
You're probably aware of the egregious cases of Daraprim[0], Acthar[1] (there are at least 5 cases from the last 10 years that I've read about, these two come to mind). They are only unique in the sense that there was no "slow boiling process", but rather an immediate extreme hike. Many other out-of-patent medicines experienced slower but significant hikes, such as the Epipen[2] are happening all over the place.
You should also read the Epipen article paragraph about the public-facing marketing, which is deceiving and life endangering, and about the policy-facing marketing, which is also not in any way compatible with your description.
It is my impression that your first paragraph is fantasy.
[0] https://www.washingtonpost.com/news/wonk/wp/2017/08/01/what-...
[1] http://www.nytimes.com/2012/12/30/business/questcor-finds-pr...
[2] https://www.nytimes.com/2017/06/04/business/angry-about-epip...
My point is that marketing expenses are far from just informative - e.g., in the case of Acthar and Epipen, they go towards convincing doctors and policy makers that a specific brand is superior or magical, when no such evidence exists; and it is my opinion (which would take me forever to find the facts for, I admit) is that a significant part of the marketing budget, perhaps upward of 50%, goes to such deceitful practices; another one that comes to mind is Oxycontin[0], which was marketed deceitfully, and is a major contributor to the opioid epidemic.
[0] https://www.theverge.com/2018/2/11/17001926/oxycontin-purdue...
I just have a very simple point to make about pharma marketing costs. It was not my argument that there aren't abusive pharma companies; clearly, there are.
In fact, oxycontins claim to fame is its uniqueness in how long it acts for and therefore reduced addiction potential, which does not actually work this way.
The “non competition” belief is itself a marketed reality for many drugs.
That's why I said "mostly". The first 10% of marketing spend may well be useful. The last 10% is the case of being louder than competitors, and is zero sum. Where the line is in a particular market varies and is subjective.