This question was written relatively quickly from a place of curiosity, sorry if I'm missing anything obvious!

When reading through some things on the debate week here, I found this comment from Arepo on why he chose to vote for global health: 

"I'm philosophically a longtermist, but suspect better evidenced short termist interventions are comparable to if not much greater than 'direct longtermism' in expectation. 

In the long run I think a thriving human descendant-line with better cooperation norms is going to lead to better total phenomenal states than reduced factory farming will."

 This comment seemed very interesting to me as I haven't seen the argument of "systems thinking" addressed that often in animal welfare or global health discussions (this might just be because I don't engage that much with the community here). What are some general arguments for and against this view? 

More specifically, a problem I've always had with certain EA metrics is that I find that they're not adressing the systems that we're inhabiting enough. It kind of feels that we're goodharting towards specific metrics of cost-effectiveness that don't capture the system behaviour?

From a systems perspective, let's consider global development versus animal welfare advocacy. Global development aims to enhance human capabilities through economic growth, education, and healthcare. The theory is that as societies advance, they naturally expand their moral circles, potentially improving animal treatment as a byproduct.

On the other hand, animal welfare advocacy directly targets animal treatment through lobbying and norm-shifting initiatives. This approach aims to expand our moral consideration for animals more rapidly and explicitly.

Both strategies have systemic ripple effects. Global development might lead to better-informed citizens making improved decisions across all aspects of society, including animal treatment. Animal welfare advocacy could accelerate the expansion of our moral circle, potentially catalyzing broader ethical considerations.

My question is: Which approach creates more significant positive change in the long run?

I'm very curious to hear what people have to say here! 

Following are some more questions that I have:

How might one capture system behaviour in metrics? 
How are we already doing this in animal welfare and global development? 
What is the relationship between cost-effectiveness metrics and system metrics?