Should moderator elections be automatically held? (e.g. at least once every X years? when there are < Y mods? etc.)

I think at the core, there's a bit of a misunderstanding of what purpose mods are intended to serve... but I want to do that while recognizing that the intention and the reality don't always align.

I've spent probably more time than most thinking about this as it relates to the SE moderation format. I think there certainly are arguments for regular mod elections but much of what you're pointing out shows a shortage of high-rep curator-type users (and failure by the company to adequately support curation).

Curators are users who spend a good amount of their time supporting existing posts on the site and making the site easier to use. On the SE network, these features are generally available to anyone with enough reputation rather than being limited to moderators. Yes, mods can do them more easily (unilaterally) but most can be done by a group of users.

Here's the list you posted, my initial reaction and the core problem leading to the issue:

• spam accounts - while it's true that only mods can delete accounts, the posts they create can be deleted with sufficient user flags. The accounts do not need to be deleted and doing so is too labor intensive to relegate it to mods, even if there are more of them.
  • Core problem: the company needs to actually intervene and do more to prevent and handle these accounts and their posts rather than mods or users needing to act. Mods have been trying to get attention to this for years. The spam problem just keeps getting worse.
• unmerged duplicate tags - with few exceptions, high rep users can do this without mods. There are a minority of cases of tag curation that require mod intervention (e.g pluralizing).
  • Core problem: the tag synonym tool is too hidden and nothing draws attention to the merge queue (not even for mods). It's essentially necessary to post about it on meta to get any action on a suggested tag merge. Changing elections isn't the solution here. You could elect dozens of mods and still have suggested tag merges that no one knows about.
• long review turnaround times - Mods are actually discouraged in many ways from reviewing when they become mods, both by the community and the platform. The intention of the platform is for mods to handle exceptional cases and reviewing rarely falls into that category. Mods have unilateral reviewing, so many mods are hesitant to review for closure unless the case is particularly clear or a few other users have already voted to close.
  • Core problems: insufficient reviewers coupled with the site settings requiring more reviews than the site can reasonably manage. This has been investigated in the past. Some have argued that mods should have the option to cast non-binding reviews but this wouldn't necessarily lead to better outcomes in your situation. Plus, knowing how to review can be confusing and isn't rewarding for most people, so it tends to be avoided or forgotten by many users.
• limits even trying to add to review queues - Mods can't control the suggested edit review queue size and increasing it just kicks the can down the road. The core issue is the same as the prior item.

The site is complex and there are lots of issues but I don't see how more elections/more mods would actually solve the specific issues you're citing. The platform needs serious overhauls in these categories and your frustration is exactly why community members get so bothered by the company investing heavily in AI but not in addressing these core problems.

I notice spam and unmerged duplicate tags, encounter long review turnaround times...

If the community isn't flagging spam or clearing review queues, more moderators (assuming you can get enough people to volunteer) is a bandaid at best, and in some cases, might cause the most active users to reduce the reviews they participate in because they now have binding votes (as Catija pointed out).

Moderators are not supposed to be super-users.

Having an election just because a site hasn't had one in a while doesn't seem like it has any benefit over just allowing a community to request an election if it feels it needs one.

I'll ask a slightly deeper question: do you think these sites are satisfied with their mod leadership? I'm not saying this to put those particular folks on the spot, but to name the implication clearly.

I want to be clear, I think it's good to speculate on alternatives / what we could be doing differently. This is especially true re elections, which really were built for a different time. I also think it's wise to build two steps ahead and put effective systems in place before problems emerge. So I'll say that I do actually believe there can be value in a periodic confirmation vote, but it's not without its costs and tradeoffs, and those must be respected, too.

The primary value of such procedures is typically in offering the opportunity to amend moderation and leadership teams that aren't working the way the community wants - whether by replacement, addition, or rarely subtraction. At the same time, while stagnancy is no good, neither is needless paperwork. (How's that for a triple negative?) If communities are still broadly happy with their moderators, and their moderators are still successfully moderating the community, then there stands to be little value added and a significant amount to be lost in trust, time, and paperwork. Hence the question above.

There's value in exploring the question further, but it's all speculative, and to become a concrete proposal it needs to leave the realm of wondering. What I'd ask you to do - if you want to pursue this further - is to clearly and blamelessly identify the problem this fixes, and to go further to explain why fixing the problem is worth the cost of the particular solution you want to propose.

No. There should not be a network-wide policy making periodic elections compulsory on every site. However, each site should be able to decide whether and how often they want to hold them.

The answer might look the same as VLAZ's answer, but the reason is different.

Below are some facts, some of which have been mentioned in comments to the question and other answers.

1. Elections take a lot of time and effort
2. Historically, despite efforts to run an election, several have been canceled because there weren't enough candidates.
3. Sometimes, when the CMs have "tested the waters" to see if the Community is interested in having an election, the community hasn't shown interest, or the site circumstances were not appropriate for an election at that time.
4. When the Pro-Tempore and Elected Community Moderators were asked if adding more moderators was necessary, they replied that it was not.

The alternative to having more Community Moderators without burning the CM and the Community is to appoint Pro-Temporare Moderators, but this also requires having trusted users willing to fill the position.

If you think a site would benefit from having an election and if they should be periodic, instead of posting here on Meta Stack Exchange, post on the Meta corresponding to that site or start a conversation about this on the corresponding main chat room.

I suspect the primary value in 'regular' fixed moderation elections would simply be that the company would have to dedicate (more?) resources to it.

... haven't held elections since at least 2020:

Isn't a great metric. We get asked every so often if we need more moderators, and Super User dosen't need more for now. This would change if one of our more active moderators left but least for now, we can manage with 'losing' a moderator for a week or two. We might want to look at it at some point but - the lack of elections isn't really due to a lack of care, it's due to the lack of need.

Admittedly our statistics are a little skewed by spam handling (and low complexity flags), but we have ~5 minute flag handling time on Super User, We deal/respond to meta posts pretty promptly - essentially, we do alright right now.

Moderation seemingly helps manage issues like spam accounts, unhandled duplicate tags, long review turnaround times, and limits even trying to add to review queues (especially askubuntu), etc.

Many of these things don't scale linearly, nor do they 'really' rely on having more moderators to deal with.

• Spam accounts

Super User currently gets 1/3 or half the spam on the network. Anecdotally speaking, a single mod can deal with most of them over the day. We do have additional reporting methods and my 'tooling' for this is a little better but having mods available at the right time (that's to say geographic diversity) has helped us more than quantity of mods.

• unmerged duplicate tag

Catija covers most of the points I wanted to make in her answer. Practically this is a design/visibility issue + more per site meta usage would help.

long review turnaround times, and limits even trying to add to review queues

Isn't a moderator issue at all - it's a sign of either a community that grew too fast for its tools to catch up, or a community that's declined enough that the high rep users aren't spending time on the review queues. And a community without enough people involved in curation and metamoderation are going to have trouble finding moderator candidates.

No

I do not see any need for mandatory periodic elections.

If elections should be held, that should be driven by the sites needing more moderators. An arbitrary length of time is orthogonal to that.

Give users a chance to feel represented

I think mod elections should be repeated based on the average user retention time of each community. This would ensure that a significant percentage of users feels represented by the current moderator team - which is in my opinion important if users should trust them to do sometimes hard decisions like suspending users, which directly impacts the community. Having elections does not mean that the mods should necessarily change, if a community trusts them, the same mods can be re-elect.