Sarah Eustis-Guthrie


Co-founder of Maternal Health Initiative, a family planning organization incubated through Charity Entrepreneurship. Writer of Beyond Denial. Formerly worked on policy in the California civil service. 


This is a shockingly good website for a...hmmm... April 1st initiative. Looking forward to your impact, great work team!

It seems great! Also their program trains grantmakers who are currently working at foundations (though not the absolute biggest ones, to my understanding) so that's even better

Thanks for the kind words! This is a great question. There's a lot of uncertainty here but here's a few thoughts:

  1. Postpartum insusceptibility is likely a decent chunk of it -- i.e. some women have "redundant" protection where they were already 100% protected from pregnancy, so the modern contraception does nothing in the short term, or they were mostly protected and the modern contraception only adds a small marginal benefit
  2. For some of the studies, there was a higher contraceptive uptake at 12 months postpartum in the intervention group than the control but the contraceptive uptake in intervention and control groups was similar at 2 years postpartum -- it seems that some PPFP interventions are primarily shifting contraceptive uptake sooner, where postpartum insusceptibility is a particularly strong factor.
  3. Behavior related to family planning is really complex and can be unpredictable/unexpected. For example, the women who  start using contraception because of a program are likely not a representative sample of the population; in fact, they may be skewed towards people who are at a lower risk of getting pregnant, because they're more careful in other ways. 

We will be publishing a journal article with our pilot findings that goes into all of the data and has much more advanced statistical analysis -- we'll be sure to share that here as well! We're also planning to publish a commentary that focuses on our broader concerns on postpartum family planning and digs into that data. I totally agree that what we've shared here is just a small sliver of the data.

I think this is a really important area, and one I've been doing some research / thinking about on my own as well. Two recommendations: CE/AIM has a grantmaker training program that does great work in this space. This also reminds me of a SSIR piece by Ian David Moss about expanding the remit of EA: 

It’s terrible to see how people have suffered due to harassment and abuse in the community. I think this is an important time for us to reflect as a community on what we need to be doing differently. 

Some aspects of the problem are more easily tractable (clearer policies on reporting misconduct at orgs; better systems for responding to misconduct), while others stem from aspects of EA that are pretty deeply-rooted (centralization of power; blurry work/life boundaries and a high level of romantic/professional entanglements). Many people seem to be pretty bought into the more entrenched aspects I mentioned, but I feel that their downsides haven’t been sufficiently accounted for. At the very least, I think we need to more robustly account for their risks, and factor them into community norms and behaviors. 

If you’re harassed by someone who controls your funding, and they’re also a beloved community member with high status, it’s going to be inherently so much harder to speak up. Yet many EA orgs have harassment policies that are poor or non-existent. Multiple people have told me that they feel such policies aren't necessary in EA because this is a high-trust community with well-intentioned people. I'd argue that when personal and professional lives are so entangled, strong policies are more, rather than less important. 

I'd be eager to hear what other actionable changes people feel would be valuable.

I disagree. IMO, many of the issues that EA faces when it comes to sexual harassment/abuse stem from aspects that are particular to the community itself. I did research for a book on this topic, and sexual misconduct and abuse thrive in contexts where power is more concentrated and there is less accountability; basically, the harder it is to speak up about someone, the more likely that their bad behavior will go unchecked and they will continue hurting people.

EA (particularly Bay Area EA?) tends to concentration of power among particular figures. And concentration of many kinds of power, including control over funding and job opportunities as well as things like social status. EA can also be pretty insular. If speaking up about someone means endangering your job and your friends, it’s harder to speak up. That’s not even getting to the fact that you might be endangering your housing situation, or might be worried about how it might affect your impact on the world. 

These factors are not totally unique to EA -- I spent a long time in the classical music world as well, and concentrations of power let countless bad actors off the hook. But I would say that they're particularly severe within EA, and understanding the particular factors that worsen things allows for more targeted solutions.

I think individuals and institutions in EA need to do a better job of mitigating risks created by unequal power dynamics. In a previous job, I conducted research related to institutional accountability and sexual assault. One common theme is that the way that institutions and communities respond to bad behavior by key figures is shaped by their norms and systems, with certain attributes making accountability more difficult to achieve. In my opinion, there are several aspects of the EA movement as it currently exists—including the blurrier work/life boundaries for many folks, the outsize power of certain community leaders, the frequent reliance on ad-hoc rather than formal systems, and the movement’s small size—that make accountability particularly difficult, and I don’t feel that we have done enough to create systems that respond to these risks. 

Let’s think through an example. (To be clear, this is entirely hypothetical.) Imagine that a woman is harassed by a prominent community leader. She works for a small EA org. Her boss is close friends with her harasser, and the org receives significant funding from his organization. She wants to say something, but she doesn’t want to threaten her job or their funding. Not only that, but most of her friends are in EA circles, and she knows speaking up would be divisive.

Some of the things that make this sort of situation more difficult in EA are based on parts of the community that would be difficult or undesirable to change. But some of them are worth changing, and the existence of all of them makes the creation of robust systems even more important. 

I think it’s useful for institutions to think through these sorts of exercises. What if a major donor was engaging in bad behavior? An organization’s leader? To what extent would victims feel able to come forward? How likely would it be that the victim would face negative consequences from speaking up, vs. that the perpetrator would face real consequences?

There are always going to be bad actors. It’s up to communities and institutions to set up systems so that when improper behavior occurs (whether harassment, assault, etc.), it is more likely that bad actors will face accountability for their actions. With rare exceptions, the deck is stacked against the victim and towards the perpetrator; good systems can help reduce how strongly the deck is stacked.

What do these good systems look like? 

  • Well-publicized, accessible systems (within orgs, community spaces, and events) that allow people to report incidents of improper behavior
  • Clear policies for how institutions will respond to reports, including how they will maintain confidentiality
  • Thoughtful procedures for reducing the likelihood of retaliation
  • Explicit conflict of interest policies for orgs and grantmakers
  • Robust governance systems

This is just a start, but hopefully a helpful one! These are conversations worth having. 

I’ve been involved with EA since 2015. I think there’s a lot of room for EA to do better when it comes to inclusivity, especially regarding gender (but also race/class/other identity aspects). 

The gender skew in EA exacerbates a lot of the issues related to gender. The gender ratio varies a ton across different geographies and cause areas, but in my experience it ranges from roughly 50/50 to overwhelmingly male (70/30 male/female per 2020 EA survey). When I walk into a meetup and I’m the only woman there, that affects my experience. This was particularly the case when I was first getting involved with EA as a teenager: part of deciding whether you stay involved with a community is your answer to “am I welcome here? Is this community for people like me?,” and repeatedly having experiences where I was one of the only women present gave me the sense that this community wasn’t for people like me. That led me to engage less with EA, though I eventually returned; I suspect it’s more common for women and people from underrepresented groups to “bounce off” of EA like this. It is genuinely surprising in many ways that EA doesn’t have more women, as women tend to be way more involved in the non-profit sector more broadly. It doesn’t have to be this way. 

But the gender skew also affects things when there are issues. If someone makes some comment that makes you uncomfortable and the rest of your male conversational partners laugh it off, that’s not super helpful. I think that, as a community, we should work to reduce the gender skew—through making EA spaces more welcoming to women, investing in mentorship programs, etc.—and actively take efforts to mitigate issues created by the gender skew. On a macro level, fewer women in the room when decision-making is occurring means that issues that affect women are less likely to receive their appropriate attention. That necessitates that institutions make a more active effort to pay attention to issues that effect women, collect women’s opinions on issues that affect things, and yes, have more women in the room when decisions are happening. On a micro level, note when you’re at a get-together and it’s overwhelmingly monolithic (in terms of race, gender, etc.). Pay attention to how that affects how you treat people in the non-dominant group. 

(splitting into a second comment b/c of length)

Could you talk a little about how selections will be chosen? Will there be a peer review process?

Load more