In what way would it be harder for them to be dethroned exactly?
Yeah, sure, the data maybe still gets through. But now it's getting through at a reduced rate and (potentially) lower quality of service. This prevents new businesses from being able to enter into many markets. Having to establish peering arrangements with many ISPs individually instead of just with the few that they directly use.
Want an alternative to AWS/Azure? Good luck, not only do they need the physical infrastructure for hosting, they now need individual peering arrangements with many ISPs to provide the same QoS.
Then with broadband you end up with users given access to a larger portion of the pool, akin to overbooking an airline. If everyone in my town gets 1Mbps but the local ISP can only handle 1Gbps, well, 1000 users maxing it out will shut the rest of the town out. This gets to a legitimate case for traffic shaping, ensuring equal access to the common resource pool. Those users can still max it, but only when everyone else isn't using it (or using it minimally).
However, traffic shaping (and related things like blatant packet dropping) started being more generally applied, to whole protocols (bittorrent) or services (Netflix, et al.). The latter example is a particularly good example of a conflict of interest. Companies like Comcast control so many user experiences in the US, but they own NBC and other media brands, as well as providing video-on-demand services as part of their cable offerings. They have a clear reason to want to reduce the QoS of Netflix and can, legally, without network neutrality. This is directly harmful to customers experiences, and a bordering on an abuse of market position.
===
Really, though, it's silly to look at how things worked 30 years ago and think the exact same rules will apply today. The circumstances have all changed. Broadband-requiring, consumer-oriented services didn't exist 30 years ago. They do today. ISPs today are screwing over consumers and prospective new businesses. They are the robber barons of the digital age, imposing their taxes at each bend in the river and screwing over everyone upstream and downstream to make a buck.
Previously, you had a lot more ISPs, even more operating in the same physical area when we had dial-up. They needed to cooperate. Their customers' desired data wasn't on their network, and they were barely more than an interlink (services offered: dial-up, email, custom home page at isp.net/~user). If they didn't abide by peering arrangements businesses and customers would have been unable to communicate, making the ISPs unable to fulfill their purchases.
There are now far fewer ISPs and they have far more services and customers. They're in a position, due to the number of customers they have to bargain with, to demand quite a bit from other ISPs and content/service providers. Failing to peer with Comcast could mean losing half of all US broadband customers. So when they demand $$, you're in a bind and almost have to pay, or suffer lower QoS as a result.
Net Neutrality is not a magic cure all and it isn't neutral when only one side has to change. both providers of bandwidth and providers of services would need to adhere to specific rules for it to not stagnate and recreate AT&T/etc we suffered through in the early days (remember Long Distance where it was all protected ... yeah, no thank you
Are you saying that net neutrality protects existing players? Let's imagine removing net neutrality for a second. Now companies can pay ISPs for preferential treatment. Who has the money to pay the ISPs? Existing players. Thus, existing players get even harder to take down
2) I think will only continue to be true for a short period of time. With products like Netflix and amazon, and with traditional television channels like HBO and ESPN starting to charge for streaming online, I think consumer attitude will shift as they realize that it's worth paying for content on the internet.
3) I'm skeptical is actually true and I'm skeptical that it's bad if true.
4) for most internet products the barrier of entry is smaller. You're only looking at one company. Compare to a company like Boeing, where it would take hundereds of millions to compete.