And it is a fiction. Here's a thorough takedown: https://hbr.org/2017/05/managing-for-the-long-term
This makes a lot of good points, but one I find particularly telling is that maximizing shareholder value is not even a well-defined goal, because it doesn't specify a time frame. A public company could be run in such a way as to cause wild swings in the share price, and the day traders and HFTers would love it, because surfing volatility is how they make their money. Of course no one thinks that those are the shareholders whose value should be maximized, but someone who accumulates a position over a couple of months betting on the price going up after the next quarterly earnings report, and who plans to sell shortly thereafter, really isn't that different — and yet somehow these are the people who have managed to seize the narrative that they are the shareholders whose value maximization counts.
And the consequences of people believing this are just massive. Offshoring, reductions in R&D, stock buybacks, a general failure to invest in the future — all of these are driven, to some extent, by this idea that the stock price must be goosed at all costs.