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!
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.
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:
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:
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!
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)