The problem with the excellence strategy is that cost is predictable but excellence is not.
AMC didn't invent the strategy of going for the high end, HBO did, and lots of networks have tried the same strategy since. AMC has done well by it, but others have not -- see Showtime and Starz for examples. This despite the quality of their programming; Showtime's Dexter is well-regarded, for instance, but has failed to give a "halo" to the rest of the network. Starz has been trying to move up-market with shows like Boss and Magic City, but for whatever reason none of those shows ever really found an audience. (Even AMC's efforts have been fitful -- for every Mad Men or Breaking Bad there's a Rubicon and Hell on Wheels.)
So betting on "excellence" is a tough thing to do, because you have to spend a ton of money without any guarantees that the public will like the results. But the results of cost-cutting are very predictable, in the near term at least; it takes a long time for a run of really bad programming to tarnish a network so much that people stop tuning in. So for the bean-counters the "logical" choice is obvious.
When people tell the story of great content they tend to focus on the very high points. "Breaking Bad" is much better than "Dexter". Even "Mad Men" which isn't as good as "Breaking Bad" is much better than "Dexter", which is arguable the best show on Showtime. Aiming for decent content isn't as meaningful as actually getting that content. Starz and Showtime have failed to get anything approaching HBO or AMC.
You can change your approach but you can't change your taste and if whoever is making content decisions at Starz or Showtime continues to have their jobs then it will not get them very far.
You see it now with the results of Microsoft copying Apple's strategy. Microsoft is doing the same thing that Apple did but since they're out of their element it's not going well. They would have been better off just being a boring company that made a great Office product and an increasingly 3rd party hardware partner friendly operating system. In the same way it would have been better for Showtime to just show movies that no one wants to see but pay for the opportunity to not watch.
You don't have to spend a ton of money. Many of the top shows such as Breaking Bad are not that expensive, and after the pilot you can always walk away and cut your losses if it's an obvious failure. There are many stakeholders in the show to share your risk too. Every show is basically a little startup. People check out the script, say fine, and then do whatever is needed. They can own shares, they can get external funding, they can even finance it themselves (it's always sunny in philadelphia).
Dexter's inability to provide a halo to the Showtime network is mostly due to the sharp drop off in quality in the middle of its run and Showtime's unwillingness to kill it there after.
Homeland might be able to achieve what Dexter failed to, but that'll be contingent on how the fix the serious missteps they made last season.
You have a very strong point. Also, excellence can't be conjured into being just by throwing money at people and hoping they have the talent to do something useful for it. (Look at the VC-funded world; if the VCs could find a thousand people like me instead of the hucksters who currently run the show, they would. But it's just fundamentally hard to find great people.) The risks and especially the search costs involved in an excellence-based strategy are high and might be intolerable from an executive standpoint.
The good news is that what I said above is almost certainly right for a single actor. When you have this legacy oligopoly that has been treating some X as a commodity-- cutting costs, letting quality fall to the most manageable level (usually mediocrity)-- then, as that oligopoly loses hold due to technological change, a single player that takes an excellence strategy will have a good chance of making a big win, because the costs of excellence are not that much higher (a lot of the best shows, e.g. The Wire, started with no-name but talented actors) and the revenue potential is 10-100x (as with programming).
The bad news is that there may be a game-theoretic limit on the rewards of excellence. The fact that it works for one or a few players (e.g. AMC, HBO) might not mean that it scales. It might be that 5 players on an excellence-strategy dilute the benefits but pay the same costs. If that's the case, then this phenomenon is no better or worse than the 20th-century market's reward for branding. (While we associate the corporate-- McDonalds hamburgers, Hershey chocolate, Starbucks coffee-- withe mediocrity the truth is that most of these products were in the upper-middle tier of quality when introduced. They decline in relative terms because cheap, available, upper-middle-tier products kill everything below them and leave themselves at the bottom.) It may not be, at least specifically to TV, that an excellence strategy works for more than a small number of players (who'd become a new oligopoly).
However, I'm optimistic because the trend for software and technology is that the rewards for excellence strategies are nearly limitless (the scarcity of excellent people and ideas, not the "room" for them, is the limiting factor) and everything, including TV, is becoming a lot more technological over time.
AMC didn't invent the strategy of going for the high end, HBO did, and lots of networks have tried the same strategy since. AMC has done well by it, but others have not -- see Showtime and Starz for examples. This despite the quality of their programming; Showtime's Dexter is well-regarded, for instance, but has failed to give a "halo" to the rest of the network. Starz has been trying to move up-market with shows like Boss and Magic City, but for whatever reason none of those shows ever really found an audience. (Even AMC's efforts have been fitful -- for every Mad Men or Breaking Bad there's a Rubicon and Hell on Wheels.)
So betting on "excellence" is a tough thing to do, because you have to spend a ton of money without any guarantees that the public will like the results. But the results of cost-cutting are very predictable, in the near term at least; it takes a long time for a run of really bad programming to tarnish a network so much that people stop tuning in. So for the bean-counters the "logical" choice is obvious.