Once there is only one paper towel manufacturer left, what is to prevent it from raising prices?
Admittedly those are public utilities but the attitude seems to hold true in antitrust as well. Walmart is probably the best example there, or now Amazon as evidenced by TFA.
Sometimes I wonder why Walmart and Comcast are allowed to behave this way while T-Mobile is not. (EDIT: Google says “nevermind”: https://www.nytimes.com/2019/12/19/technology/sprint-t-mobil...)
As a super-obvious example, an accountant looking to cut costs at a hamburger chain might first suggest reducing the amount of meat or using old meat. But that reduces value as much or more than costs, so it's a bad optimization.
That depends on the significance of economies of scale and barriers to entry in a particular market. The term “natural monopoly” (as it’s used in economics) refers to a particular market where, because of barriers to entry, the optimal number of firms is one. Two firms would not be able to produce their good for cheaper than one firm.
I also suspect the notion of "natural monopoly" is oversold and too simple. Would it be more efficient if we had exactly one ISP for the country? In theory, yes, because then we only have to run one set of wires everywhere, and we'd get rid of a lot of duplicative equipment and staff. But in practice, monopoly and oligopoly ISPs are generally both expensive and bad. I just moved from a competitive area to a "natural monopoly" area; my internet now costs twice as much for 10% of the bandwidth, much lower quality, and much worse service.
I think that's because companies aren't static entities that reliably produce goods, even though that's what most people imagine. Instead they're temporary coalitions of individual actors hopefully prodded into optimal behavior by external forces like competition. Especially so given American business culture, which often refuses to recognize ways of thinking that might mitigate the problems.
But if the reward for success is just having Amazon come in and hoover up the money you would have gotten by launching a knockoff, then suddenly there's a lot less incentive to invest in novel products. That's true both for categories where Amazon is competing and ones where it isn't currently.
I wouldn't be surprised if Amazon's replacement product is sometimes modestly worse, because a) they don't have the kind of deep expertise in a product that the original creators do, and b) it doesn't have to be as good to get the money.
And then there's after-sale support. Amazon's customer support is atrocious. The one thing they're good at is taking things back. But anything more complex and it's a nightmare.