The following is taken mostly from my answer at How to reasonably scope a list question asking for anime of humor genre?, though I've incorporated some ideas from atlantiza's and coleopterist's answers on What is the status of list questions on this site? as well. In some cases I've quoted word-for-word from my own writing without indicating the original source (the answer linked above), to preserve the flow of the writing and limit the amount of new text needed to compose this answer.
The main difficulty with our policy on list questions in my view is #2, though the other points also have some problems. But #2 is the most problematic in my view. Here it is:
The list items can be determined objectively. Don't ask about a list of our favorite things, or things we think are "good".
What's the trouble with this, you ask? One is that it isn't exactly clear what constitutes "objective". But the worse problem is that it isn't strong enough. What we actually want isn't that the individual items of the list can be determined objectively. That's fairly easy to do by defining terms sufficiently well. But even if that's the case, we can still run into cases where people post different lists, each of which contains many of the examples of whatever is being asked about, but not all of them. Then it again becomes a subjective "which list do I think is better?" type of question.
Rather than imposing criteria on the individual elements of the list, we should be imposing the objectivity criterion on the list as a whole. That is to say, we should require that the lists are arguably complete, and that no more examples beyond what is listed exist. This is a much stronger condition in practice, and it's the only way these questions can avoid having the exact same problems of subjectivity and list maintenance as recommendation questions.
However, it's not a viable policy in practice. The difficulty is that very few lists are likely to satisfy it, especially if we're talking about out-of-universe lists. The only exceptions are things like "What are all the movies that Hayao Miyazaki has directed?" which tend to not be very interesting. The difficulty is that you necessarily need another source to confirm that the list is complete. There are very few sources like that, and all of them are well-known (e.g. in this case, Wikipedia). If we required this, the only sorts of answers that would be possible are essentially link-only answers. What's more, an external reference, while a good argument that an answer is complete, can still have errors. So that isn't a real solution. It's also even harder than our current policy to apply uniformly.
What we need is a simple policy that can be understood by everyone and clearly applied. The current policy on list questions allows some good questions, but it becomes a point of debate whenever we have anything controversial. Almost all of these controversial questions are asking for lists of anime/manga, and almost all questions asking for lists of anime is controversial. Hence, seeing no better alternative, I propose the following policy change:
Questions asking for lists out-of-universe, e.g. lists of anime, manga, etc. are banned. This includes questions asking for examples of a particular genre/theme/trope.
This is the policy on Scifi and they don't seem to have problems with it. Arqade has a similar policy as well. I could not find any graduated SE sites that would allow questions analogous to our questions asking for lists of anime. In fact, I also couldn't find any beta sites which would have allowed these. Admittedly I didn't look terribly hard, but I still think the point stands.
This is very simple to apply, though there are a few cases that you have to be careful. For instance, Has an anime ever come out before the manga was published? is not a list question, even though the answers contain lists of anime. It's merely asking if it's possible to have an anime before a manga, and so it's a question about anime production. It would be possible for someone to come up with a totally theoretical answer to the question addressing all of the issues associated with producing original anime, but without any examples. This question should maybe be protected to prevent further answers that just list examples, but it isn't a list question so it should be left open. That's just how the answerers chose to answer the question.
Contrast that to https://anime.stackexchange.com/questions/2645/im-looking-for-a-list-of-humorous-anime-similar-to-oruchuban-ebichu. There's no reasonable way to change the question into something that doesn't violate this policy. There isn't any minor modification that could be made to comply with this policy. So we would not allow that question. This policy would also affect some other older questions which were reopened after the original list question proposal.
It's worth looking at the questions which we'd be losing if we adopted this. After browsing the ones on the main site, none were questions that I'd be particularly upset to see go. There was not a single question in this category, after over a month and a half of this policy, which was interesting to me. Naturally, this is subjective, but I suspect most will agree with me on this. Most of them were either bad questions or tolerable questions that weren't interesting to very many people. It's worth pointing out that this is coming from someone with, as of right now, 300 question votes (the second most on the site), so my criteria for interesting content aren't terribly strict.
My feeling is that if we've generated a half-dozen or so bad-to-mediocre questions of this type, and 0 good ones, it just isn't a good type of question. You can probably come up with special cases where such questions might actually be interesting, but no one has asked any. If they are common enough that we should allow the bad questions just so that the few good ones can also have a home here, then why haven't any of these been asked? In any case, making a policy based on questions that have not been asked, when we have a problem with what is being asked, is a bad idea.
Furthermore, as I mentioned above, questions of this type tend to already be answered at one of the standard references. No one is going to come here asking "What are all the movies Miyazaki directed?". They'll use Google, or go to Wikipedia, or some anime-specific resource. Losing these hypothetical "good" questions which are actually trivial isn't something we should worry about, and it's certainly not worth conflicts with new users over confusing policy.
tl;dr These sorts of questions aren't very interesting, and they lead to debate even among high rep users. Rather than allowing them in certain complicated situations, we should ban them altogether to avoid conflicts of opinion and simplify the rules.
Note also that these sorts of things are great for the Recommendations and Reviews chatroom. If a user wants to ask these sorts of questions, I'd advise directing them there.