Re: Why would a reseller police positive reviews?
That's a fair question and I'll try to address it from the profit motive.
Because Amazon has reached such high volumes. They are the top dog in online retailing so they have the furthest to fall.
Because online retailing is all about trust and tolerating obvious leeches is a reputational risk.
Because Amazon has a huge stock market valuation and even higher P/E ratio. That P/E is justified by the belief that Amazon can extend into more offerings. Few new Amazon ventures will be safe from a perception of sleaze tolerance.
Amazon has been hit in the press before, from the perspective of how easy it is to conduct transactions to buy reviews. And they are taking the review brokers to court. But in this case, Amazon came off as the aggrieved party. They couldn't exactly know exactly what was fake and I will be the first to say that the risk of nuking real reviews by mistake is a good defense for not just going medieval on anything that smells like a rat (like the guy further down who says he disregards anything above or below a 3).
However, with obvious, repeatedly reported upon, fakers that are still operational, Amazon is a lot more visibly in the know. The products in case are, objectively, scams. The reviewers are all shills and visibly so. Customers have complained about it, been acknowledged, posted about it and you still see the scammers. And Amazon's own cross-referencing system makes it easy to follow the sleaze. If you go to the "Customers Also Bought Items By" on the author's page, you can see the other vendors that are using this pool of the same reviewers and drilling into those products brings up the same disgruntled 1 star reviewers asking why this is allowed to go on.
This is not in need of some fancy undercover investigation on Fiverr.com. Out of a court of law's burden of proof, this is damning evidence of neglect by Amazon.
To put it differently, if, like your question is asking, the neglect is due to a profit motive, rather than say legal uncertainties about how to proceed, then this is a bit like Lenovo's recent problems. How much money are they making from low volume, but obvious scammers, versus how much do they have at risk if it blows up in their face?
Benefit to risk ratio seems overwhelmingly to favor at least getting to plausible deniability, which means shutting down obviously cheating vendors. Purely from a tooth and claw capitalist point of view, not altruism in the least bit.