I think politics can seem very opaque, incomprehensible, and lacking clear positive payoffs but after volunteering, studying, and working on campaigns for a few years, I think it's more simple but difficult. 

1.  I think politics is an area where there are a lot of entrenched ways of doing things as well as a lot of pitfalls that often require experience to navigate well. And even then, the chance of failure is still high. A moment's slip up or bad assumptions or a random event can undermine months or years of work. This doesn't happen as often in other areas. 
2. For animal welfare, I think the outcomes show that it's something people are more willing to vote for than pay for, so I think ballot initiatives are generally a good route to try out. I think the pork industry challenge to the MA law is pretty weak, but even if the initiative is struck down, it was probably good to try and see if it worked, and that may still open up some new opportunity. Winning by a large margin is good in that it may discourage special interests from trying to run a counter ballot initiative next time to repeal it.  
3. I think it's important not to become naive about anyone elected to office. Just because they have a similar background to you, say things you agree with, or belong to a group you like doesn't mean they're going to actually do good things or that the things they do are good. Just because they seem right about one or even many topics doesn't mean they know what they are doing on other topics. 
4. Politics is about coalition building and that often means various kinds of deal making. This is not for everyone, and not every deal is good or even necessarily clearly good or bad. It also involves constant tradeoffs and high uncertainty that will often make a lot of people unhappy. 
5. Politicians spend most of their careers fundraising - even when in office - and not nearly enough time talking to groups of their constituents that represent the diversity of experiences in their districts. This means a lot of popular ideas get ignored, some of which are good and others which are maybe not. Being a good representative means knowing when, how, and how much to defer to people. 

Holly, thanks for posting this. You raise important and interesting questions.

I do think EA should continue to steer clear of overtly partisan-political issues, especially in highly polarized societies like the U.S. This is for a few reasons:

1. EAs seem to lean Left politically, according to EA surveys; but to engage with partisan politics in a morally and epistemically humble way, we'd need a real diversity of political attitudes within EA to provide checks and balances against unconscious political biases and virtue-signaling.
2. Partisan politics tends to involve controversies that are often not neglected, tractable, or scope-sensitive, so they're often a poor fit for EA. Modern political parties exist not to solve real-world problems, but to raise money, get votes, amass power, support virtue-signaling, and make partisans feel good. EA exists to solve real-world problems efficiently and open-mindedly, so its principles and goals simply 'don't compute' to most partisan politicians, activists, and voters.
3. Politics is a game for older, cynical sociopaths who understand how power works, not for idealistic young EAs who, dare I say it, are often quite naive about the nature of power and how it's gained and wielded. (I'm in my mid-50s, and I'm only just starting to understand how American politics actually operates; I'm dubious that EAs younger than about 40 would end up being anything more than pawns in political games they don't understand.)
4. Political engagement is highly addictive, rewarding, thrilling, and distracting. (Trust me on this, as someone who's all too engaged on political Twitter.) For many EAs  trying to work at the intersection of politics and EA, the political side would often end up pulling them away from their EA principles and cause areas.
5. Politics tends to be very nation-specific and culture-specific, whereas EA aspires to global relevance. Insofar as EAs tend to be from the US, UK, Germany, Australia, and few other 'Western liberal democracies', we might end up focusing too much on the kinds of political institutions and issues typical of these countries. This would lead to neglect of other countries with other political values and issues. But even worse, it might lead us to neglect geopolitically important nation-states such as China and Russia where our 'Western liberal democracy' models of politics just don't apply very well. This could lead us to neglect certain ideas and interventions that could help nudge those countries in directions that will be good for humanity long-term (e.g. minimizing global catastrophic risks from Russian nukes or Chinese AI).

For all these reasons, I think EA should strive to avoid the realm of politics, and just do our own thing, in what we could think of as 'non-political stealth mode', flying under the radar of the usual partisan-politics system of news, activism, power, and propaganda. 

It’s my impression that development NGOs accept having to do things like bribe corrupt policeman to use the roads as operating costs. This is how it is for scientists doing field work, at least, which is a world I know better. Knowing how and when to bribe is an important skill for people who travel in developing countries, and among those people it is generally viewed more like deferring to local custom and respecting that you a foreigner in their country than a sullying act of corruption.

• I think if both optics and impact both constitute a significant part of our motivations for doing politics, we shouldn't be doing it.
  • The costs of compromise are often exponential with the number of separate criteria we optimise our projects for. It's exponentially harder to find one date who's rich, kind, smart, blue-eyed, tall, whatever, compared to finding $n$ dates who are separately significant in $n$ traits. I guess I'm advocating for the cause-prioritisation equivalent of polyamory or something. Oo
  • To optimise separately for optics and impact, ask:
    • What's the most effective way we could achieve a good reputation?
    • And what's the most effective way we could achieve impact?
      • Although projects purely optimised for impact should probably still compromise with reputation as a satisficing side-constraint.
  • Within animal advocacy, perhaps it's better to separately invest in a portfolio of projects that legibly help relatable animals, and a portfolio of projects that are maximised for impact in a way that most people find hard to understand.
    • The latter, imo, looks like trying to somehow leverage the transition to a post-AGI world for the interests of animals. Or something.

Yeah there are a lot of "fairweather friends" in politics who won't feel inclined to return any favors when it matters most. The opposite of that is having a committed constituency that votes enough in elections to not be worth upsetting - aka a base of people power. These take serious effort to create and not all groups are distributed geographically the same way so some have more/easier influence than others. One reason the NRA is so powerful and not abandoned despite negative media coverage is that they have tight relationships with Republican politicians and they turn out big time in any primary where someone opposes them or something they want. It's not so much about the campaign contributions as far as I can tell (other groups spend far more and are much less influential) though campaign contributions are certainly a part of their system of carrots and sticks. 

The lack of more broad participation in primaries is a problem for represenation and responsive good government. It's an opportunity for groups that aren't all that representative to magnify their influence. Alaska's top 4 primary election seems like a step in the right direction since it opens up primaries to more voters and then  lets voters rank the top 4 candidates in November. It increases the chances that someone can try to run and win as a more representative candidate instead of being filtered out by small, highly partisan groups. 

It's often easier to stick to established narratives, group identifiers, and allies, or even make up new conspiracies than to be measured and nuanced. Something inflamatory and/or conspiratorial is more likely to hook into human brains, be amplified by engagement seeking algorithms, and, if it's obscure but rapidly repeated, not have any better sources of information competing with it when people look up its key words. 

You pose some great questions, Holly. 

I really like the EA culture of "unintended consequences" analysis, as you do with a political campaign on rodent control. That analysis is good to do with all campaigns, and some political campaigns will have greater risks and some will have lesser. It is not inherently a reason not to engage in politics.

Interesting points about undermining democracy. I think it is also worth considering the null-hypothesis: what is the impact of doing nothing? How does it favor existing powers to continue the status quo?

Ultimately, the costs and benefit of political campaigns are extremely difficult to quantify, so I understand why EAs are still developing our attitudes toward this. As a person who works in politics, I agree that there are potential consequences to be wary of (developing a partisan identity, making powerful enemies, etc). Nonetheless, I see these as important precautions and risks to mitigate rather than dealbreakers.