Hide table of contents

Summary: In a situation where your coworkers and other applicants share your value, you can have an impact where no one else would do the job (e.g. because you have rare skills or it's an undesirable job) or by doing the job better than the next-best candidate (who will then be free to do another job better than the next-best candidate and so forth). You can also be entrepreneurial at work and carve out a new niche for yourself that closely matches your unique skills.


When I'm applying for jobs, I consider many factors - how would this job build my skills for future roles? How much does it pay? Would I enjoy the work? And how much good could I do in this job?

When thinking about how much good I could do, I consider my tangible impact - the direct good and/or direct harm I would do in that job. For that reason, I would never apply to be the CEO of a tobacco company, even if I were somehow qualified and could make a lot of money doing it.

But I also think about my counterfactual impact - what would have happened if I never applied for the job. In most situations, the next-best candidate would have done the job instead of me. She would have probably had similar values and skills to me. What does this kind of replaceability mean for my career choice?

A lot of people I know in the EA community are really interested in this topic, so I decided to write out what I know about counterfactual impact when your coworkers share your values. That means what I'm writing below doesn't apply to Earning to Give careers, where the next-best candidate probably wouldn't be donating so much of their income, or to careers where the next-best candidates could have very different values from you like in party politics. This article will be most relevant to staff at a charity or civil servants working on something uncontroversially good, like improving public health.

Jobs that no one else would do

When it comes to replaceability, the easiest way to have a counterfactual impact is to do a job no one else can or will do. In these situations, your counterfactual impact will be roughly the same as your tangible impact, at least until another person could have been recruited. Finding a job no one else would do is pretty rare but there are a few situations where it's more likely:

  • You have a particularly rare and valuable skill. When a particular skill is in high demand, there are often many important jobs using that skill which go unfilled. If you're an expert in machine learning and you're willing to work in Government or a small charity, you'll probably find a job that could have gone unfilled. Even if the job you directly applied for had several applicants, you'll be freeing them up so that one of them takes a job that would've gone unfilled. For more on developing rare and valuable skills, see this 80,000 Hours interview with Cal Newport.
  • You could also be entrepreneurial and carve out a completely new niche for yourself. For example, Pardis Sabeti from Harvard, who uses computational techniques to help really make progress on some pretty serious diseases, works on the cutting edge of computational biology and found a niche for herself. People around the world have started local EA chapters for their city or university.
  • You could be in the right place at the right time. For example, if you already have Government security clearance, during a national crisis like Covid-19 you may be offered work that simply wouldn't get done if you don't do it. During an emergency, normal hiring procedures take too long, so even if you wouldn't normally be the best candidate, during an emergency you might be the only candidate and you can have a really important impact.
  • You could do a job that is particularly undesirable or has high turnover. For example, some jobs are not very prestigious or don't pay well. Some involve taking on personal risks or living far away from friends and family, like fighting the Ebola crisis in West Africa. Other jobs are very demanding, and so people regularly leave the job after a short time, but actually could be done much more effectively with someone staying consistently in the role for a period of several years. Be very careful taking on a job that no one else will do, especially if you don't understand why. This kind of job can cause physical or mental health problems that can reduce your effectiveness in the rest of your career. Get external advice not just from other EAs but also from friends and family who love you before you take on an undesirable job that receives very few applicants and/or sees high levels of turnover.

Jobs where you're the best candidate

If you've been offered a job, congratulations! You should probably assume you were the best candidate. Here are some ways you can have a positive impact compared to the counterfactual of the next-best candidate getting the job:

  • You are better at your job than the next-best candidate. Given you're the person who was hired, this is almost certainly true! There are some situations where being a little better at your job means a little bit more impact - for example, if you were hired as a fundraiser, maybe the next-best fundraiser would have raised $800,000 and you raised $1 million. But in some situations, even a small difference in skills can have a big difference: if you're talking to a new donor who decides to donate $1 million, he might not have donated any money if he had spoken to a less skilled fundraiser. In some situations, even though the skill levels are similar, the impact can be dramatically different.
  • By taking an important job that uses your skills, you are also releasing the next-best candidate to do another job where she's the best candidate. Imagine you apply to Charity A as a software developer. The next-best candidate was also good; maybe there's not much difference between you. But that candidate is now free to work at Charity B, who would've otherwise hired a less skilled developer. That less skilled but still competent developer is then free to work at Charity C, replacing someone who probably wasn't totally qualified for the job. Even though the difference between you and the next-best candidate wasn't huge, you've improved the impact of multiple organisations by adding to the supply of talented candidates with a valuable skill.
If you take a job as the most skilled applicant, you're potentially improving the impact of several organisations, not just your own.
  • You can also have a counterfactual impact in your job by coming up with new ideas that use your unique skills and perspective. For example, if you are a civil servant working on pandemic prevention strategy, you could suggest the team include both natural pandemics, lab-leak scenarios and bio-terrorism in your strategy. Even if everyone else in your team agrees with you because you all have similar values, they might not have thought of that idea because they don't listen to the 80,000 Hours podcast as much as you do. Or perhaps you've been hired as a writer for an EA organisation and not only were you the best candidate to write in English, you're also fluent in French and Spanish. These extra skills give you an opportunity have a unique counterfactual impact.


Counterfactual impact isn't the only thing I consider when applying for a job - I think about my tangible impact, how the job would set me up for doing good in the future, and how the role fits with my personal life and aspirations. But counterfactual impact is important, and as someone who works with skilled coworkers who also want to do good, sometimes I worry that I'm not having much of a counterfactual impact at all!

Thinking through counterfactual impact in careers where your coworkers have similar values has made it even more obvious to me the importance of developing rare and valuable skills and of being entrepreneurial - suggesting new options for my role based on my unique perspective and experiences.

Thinking through the counterfactual impact in different scenarios has also helped me to notice how I've had a counterfactual impact in my career. I have had an impact in my policy career by being better than the next-best candidate, by being in the right place at the right time during emergencies like Covid-19, when I've had new ideas, and by taking an important job that may have otherwise been vacant for a couple of months.

This post was written quickly and all mistakes are my own. If you think you've spotted a mistake, please comment so I can improve the article. Similarly, even after writing this, I still don't know how to compare the counterfactual impact of two different policy jobs I'm considering, so if anyone has ideas on comparing the counterfactual impact of roles where your coworkers and other applicants have similar values to you I'd be very interested in your comments. Thanks for reading!





More posts like this

Sorted by Click to highlight new comments since:

You can also have a large counterfactual impact if you free up other people to do more important things. Eg, if you're an EA lawyer who can  annually help save 4 FTE-years of filing paperwork or doing legal research. 

For example this comment by catehall was super useful for me. I'm sure that even without a legal background, I or someone from the RP ops team could have figured out the right answer eventually, but it would've taken us hours if not days, while that comment probably took catehall like 10 minutes, so she had both a large comparative advantage and absolute advantage in answering that question, relative to generalist researchers or generalist ops. 

There's a similar story for being a programmer that automates other people's work, or for being a research assistant to great researchers, or even for ops writ large.

I've seen 80,000 Hours say something similar, but I don't actually think this provides counterfactual impact unless one of the things I listed above is also true.

If you're hired as a research assistant or programmer and someone else would have done the role equally well otherwise, you wouldn't have any counterfactual impact. It's only if the role wouldn't have been filled otherwise, or the other candidates wouldn't have taken the initiative to automate others' work, that you have a counterfactual impact.

Ah right, not just releasing the next-best candidate to do another job, but helping other people save time as well (in a better way than another candidate would ie because you have rare and valuable skills)

I think the most impactful civil service job would be a job I created for myself, in an important area I've worked in a long time, using my unique perspective, institutional knowledge and rare skills.

That's very doable - when you've worked in a policy area for a while, you can often make the case that a new piece of work needs to be done and you're the person to do it.

I'm not sure how to compare that to other jobs I could do though!

Curated and popular this week
Relevant opportunities