Ever felt like something's missing from how EA talks about “doing good”? Like the conversation stops just short of asking what's actually broken at the root? This opening post introduces a sequence born from exactly that itch: a group of EA-adjacent volunteers who spent three years exploring whether there are systems-level dynamics and interventions that the community should be paying more attention to.
Post 1: In this first post of our sequence, we set the stage for the empirical and conceptual work to come: we outline who we are, how our collaboration came about, why we believe systemic change should be more of a focus area within EA and what we are hoping to achieve with this sequence. We hope you consider this as an invitation into an ongoing conversation.
In the coming posts, we will then:
Post 2: Bedazzle you with definitions – what even is “systemic change”?
In the second post of the sequence, we propose an understanding of systemic change-related work that emerged from our internal discussions. Based on that, we outline key criteria that organisations should pay attention to when designing systems-conscious strategies. We offer these as a starting point for discussion across the community and for adaptation by each individual organisation or project.
Post 3: Excite you with data – do EA organisations actually pay attention to systemic issues already?
In the third post, we describe our empirical investigation. We conducted this in the form of a questionnaire to assess the systems-orientation of existing organisations. We will go into detail on our methodology, present the results from our empirical exercise, and discuss several limitations which reduce the robustness of our findings.
Post 4: Share our joy and frustration with you – and imagine the future together.
We pick up on these limitations in the last post of the sequence. This post collects open questions, remaining uncertainties, and potential points of contention that we would love to see discussed across the EA community. We close the post with the motivation that started the whole project: our case for an EA-inspired effort to change systems rather than just addressing isolated problems.
The case for focusing on the transformation of systems rather than aiming for incremental improvements within a system can be mapped onto cause prioritisation frameworks like ITN or cost-effectiveness analyses, with a slight twist:
Tackling systemic features that lie at the root of pressing problems is highly important, because such systemic features are often huge levers for improvement. For instance, removing lead from gasoline, water sources, or paint products is a higher-leverage intervention than treating diseases and cognitive impairments after populations have been exposed to lead. Preventing new lead from entering drinking water and other everyday products through changed laws and regulations has even more potential positive impact. At the deepest systemic level, eliminating problems like lead exposure would be achieved most effectively if there were not just new rules but a shift in what societies consider acceptable risks to impose on others. That kind of shift renders it inconceivable for actors to recklessly contaminate living environments in the first place.
Tackling systemic issues successfully is moderately tractable. It is true that systems are difficult to understand and predict. Thus, effective interventions are less tractable than they would be for more well-contained, experiment-prone problems. But there are cases of tractable work to promote systemic change. To take just two examples:
Tackling systemic issues is ... possibly neglected? We as the authors of this group share an anecdotal sense that this is the case, especially within and adjacent to the EA ecosystem. Our empirical investigation (read Post 3 of the sequence, forthcoming) gestures in a similar direction: It seems like EA-adjacent organisations consider systemic issues mainly in passing and rarely as their main strategy for impact. But that investigation faced major limitations and is not robust enough to turn our results into strong claims. If you disagree on the neglectedness point, we would love to hear from you down in the comments!
All that said, we believe that the case for a stronger orientation towards systemic change also carries a benefit which is a bit awkward to fit into the ITN framework (and similar cause prioritisation frameworks): An effective change to a systemic structure is sustainable in a way that one-off interventions often cannot be. Even if interventions addressing the symptoms of a problem might perform well on a cost-effectiveness analysis in the short term, their impact might be short-lived. Problems come back when the intervention is scaled back. The flip side to that is that systemic change must be carefully evaluated to prevent sustained negative change that is hard to reverse.
Looking at systemic change with the help of these criteria (importance, tractability, neglectedness, and sustained impact) suggests that there is more room in the EA ecosystem for approaches that lead to radical change of systems. We advocate for more research and entrepreneurial attempts to further our understanding of how this could be realised and will propose a long-list of solution directions in the final post of this series.
We write these posts as private individuals, not as representatives of any of the organisations we are each affiliated with. This project was not backed by any funding and is purely based on volunteer work. We have different levels of engagement with the effective altruism community, but we all share a commitment to the basic principle of using reasoning and evidence to tackle the world’s problems effectively. We also all believe that the effective altruism community has built an impressive space to flesh out, discuss, and act on this principle.
The idea for this project emerged through many informal conversations. We had these amongst ourselves, with colleagues who collaborated on the project in its earlier phases, and with others in the EA space (at conferences and other social events). In these conversations, we noticed that we shared a sense of dissatisfaction and confusion about the lack of connections between the EA community and other political and social movements. Many such movements are motivated by similar goals: reduce suffering and increase “moral goods” in the world. But they are often informed by quite different ideas about how to achieve them. More specifically, there is one perspective that seems less prominent in EA than in other movements for doing good. That is a focus on systems dynamics and a goal of tackling problems at the systemic level.
We are aware of previous discussions about systemic change within EA (EA forum posts on the topic, 80,000 Hours). A growing number of political and lobbying-based interventions are already being pursued by EA-aligned organisations. At the same time, significant scepticism remains. Official EA commentary (see the relevant section in the FAQ on the Effective Altruism website) reflects this ambivalence: It acknowledges that systemic change can in principle be a promising avenue. But it also cautions that large systems are difficult to change without unintended consequences, and that opinions are heavily divided on whether major structural shifts would be net positive. Our impression is that especially initiatives that would push in the direction of more radical systemic transformation have continued to be neglected in the EA ecosystem. There are important questions regarding this field that EA has largely left unanswered:
Feeling uncertain about all the questions above, we concluded that there might be value in an honest attempt to encourage constructive conversations within the Effective Altruism community. We felt keenly aware of the risk for such conversations to dissolve into unhelpful infighting, where people talk past each other because they hold starkly different starting assumptions. With this sequence, we hope to help avoid this unproductive outcome.
We spent the last ~three years investigating what we mean by “systemic change”: We developed a list of criteria to rate organisations based on their level of engagement with systemic issues/considerations, applied these criteria to existing organisations in the EA ecosystem, and analysed the results. All of this aimed to produce a decent empirical overview of the status quo.
The data from our investigation suggests some differences within and across EA cause areas. However, our findings are limited due to the low reliability between raters and time-constraints on our research. Therefore, we refrain from drawing generalisations from the data and focus on a descriptive analysis instead. We discuss the limitations and what we can learn from them in the last post of the sequence.
We welcome any thoughts or questions in the comments (or via direct communication to ourselves, if preferred)! This sequence is very much meant as a conversation starter. We hope it generates engaged, constructive discussion on the importance and feasibility of systems change work. We also hope it sparks discussion on new projects to explore how systems thinking and EA principles might be brought together, or how they might clash.
So if you have any thoughts in that direction in response to our first post, please do let us (and the rest of the Forum) know below! This includes critical remarks on our underlying assumptions or the specifics of the project we introduce here.
