As I mentioned in my strategic plan, I have 10 months and 28 days until I graduate from Denison University and hopefully will be transitioning to a career.  Careers are important because having one will not only mean that I won't starve, but that I'll have an opportunity to change the world.  As the career advice organization 80,000 Hours notes you'll be spending about 80,000 hours in a career, so you might as well use it to make as big of a difference as you can?  But what career should I pick?

Here are my starting thoughts and strategy...

 

Basic Philosophy and Strategy

My aim is to choose a career that will, taken as a whole, contribute the most to the world with regard to my utilitarian goals to increase total well-being.  "Taken as a whole" means that I'm not just considering the direct good of the first career itself, but also it's interaction with the rest of my life, it's indirect good, and how it might set me up to have an even better second career.

However, there is one catch: I furthermore want the job to be at least moderately enjoyable.  I would very much not like to be depressed or burnt out, and I suspect that would be bad for my utilitarian goals too.  If I think I would be miserable at my chosen job even if somehow it were to maximize total well-being, I still would not take it (despite this being immoral from a utilitarian standpoint).

I also would like to have a plan in place rather soon, so I can actually act upon preparing for it.  The earlier I pick a career plan, the more options will remain open.  Right now, I'm stuck, for better or for worse, with a psychology and political science major and unless I do something drastic, I won't be able to take any classes outside of those two majors for the rest of my time at Denison.

As a political science and psychology double major, I think I could be very well qualified for a job that involves research and/or statistics.  To get more of an idea of what I can do, feel free to look at my résumé and at my personal website.  I'll speculate a bit more about my qualifications later on in this essay.

 

The Scope of High-Impact Careers

The way I've been categorizing it, there are essentially only two different kinds of careers that promise to do lots of good -- funneling money to effective organizations and working for effective organizations, though some opportunities allow for some combination of both.

By effective organizations, I mean orgs like Giving What We Can, 80K Hours, Effective Animal Activism, The Life You Can Save, Center for Applied Rationality, Machine Intelligence Research Institute, Future of Humanity Institute, Leverage Research, The Bill and Melinda Gates Foundation, GiveWell, Against Malaria Foundation, Vegan Outreach, etc.  (Of course, the actual choice of organization would matter a great deal, and some of these organizations might actually not be effective for one reason or another, but I won't look at those arguments here.) 

 

Funneling money to effective organizations, in this case, involves only one opportunity -- earning to give, which I've summarized before. The basic idea is that you aim to pick a high income job and then donate as much as you can to an effective organization.

Working for effective organizations involves four other opportunities -- research, influence, fundraising, and/or direct work.  I say "and/or" because you could end up employed at a effective organization that allows you to so some combination of the four.  Research might involve working for a university working on an important question.  Direct work might involve assisting in the day-to-day opportunities of the organization.

Notably, research, influence, and perhaps even fundraising, could be carried out in one's free time, such as what I do through this blog.

 

My Perceived Opportunities in Earning to Give

I think earning to give is a great baseline for me to evaluate other opportunities from, even though I'm moderately confident I can do more impact through a different line of work.  What opportunities await?  Here are my guesses:

  • Investment Banking: My guess is this is the highest earning career, when considering probability of achieving certain incomes multiplied by the quantity of income, and high incomes right out of college.  However, I'm only moderately qualified for it -- I think I have a good analytic mindset and have taken some economics and finance classes, but I'm not an economics major, and I've never done any finance internships.  Moreover, I don't think this job would be particularly enjoyable.  I see a high chance of failure to get a job and burnout once I get a job here.
  • Law: Law seems like a high-earning career if you can make it big, but there's a low chance of making it big.  I've heard anecdotally that you basically need to get into a top law school or you won't be having a good time, assuming you can pay for that law school.  I think this would be moderately enjoyable and with Moot Court, a Constitutional law class, and a Political Science major, I'm as prepared for law as anyone can be in a liberal arts school.
  • Consulting: Consulting offers high incomes right out of college, though the incomes might not be as high as other opportunities overall.  I'd say I'm as prepared for consulting as anyone can be in a liberal arts school, though I'm not an economics major.  Unfortunately, however, I'd guess that I'd really hate regular travel, so consulting might not be a good career path for me.
  • Computer Programming: Computer programming also offers high incomes right out of college, provided you're talented at it.  I think I have the raw aptitude, but I don't have the formal training.  My guess is that if I wanted to go into programming, I'd have to pursue some additional education after Denison.
  • Market Research: I think market research would also offer decently high incomes right out of college, and I'd be well qualified for it and I'd enjoy it.  This has always been one of my favorite choices.
  • Engineering: Engineering would provide high incomes right out of college, but I have no experience with engineering.
  • Medicine:  Doctors earn a lot, though they have high expenses with medical school.  Doctors seem to be like lawyers who have a more smooth earnings curve -- you'll earn a bit less overall, but you have a higher chance of success.  However, I don't think this is worth pursuing as I have no interest for it and I have never taken a single college-level biology course.

 

My Perceived Opportunities in Research

I've done really, really well in political science and think I'd have lots of aptitude as a political science Ph.D.  However, I don't think political science offers much opportunities to actually make an impact with one's research.  Psychology research, on the other hand, I do think has a wide variety of high impact questions to study, but I've only been moderately good at psychology and I'm unsure I'd get into a top Ph.D. program.  I also imagine I might have the general skills to survive in other social science programs, like economics research.

I thought about doing philosophy research, since most of the biggest questions seem to lie in the realm of philosophy right now.  However, I'm nearly certain I can do just as well or even better at philosophy research by just writing things on the internet.

 

My Perceived Opportunities in Influence/Fundraising

I don't know if there's much I could do in influence beyond what I'm already doing.  I suspect my only options are to either retire early and blog full-time (it's not that hard to retire permanently by age 30 with a high enough income and savings rate).  That seems unlikely to be my best choice.

I could go work on behalf of an effective organization, which seems promising.

Or I could always retire early and then work for an effective organization, as well, which would have the added benefit of not needing them to pay me.

A career in politics is unlikely to be high impact as a first career, and I've probably already burned myself from running for higher office with public statements of outside-the-mainstream views.

 

My Perceived Opportunities in Direct Work

I think this is the same as the above.  I think I'd enjoy working for an effective organization directly.

One interesting idea is to be a personal assistant to someone who is high-impact to boost their impact.  I'm not sure how good I am at this, however, and I'd feel like I could do better.  I'm also not sure if this would be fun, though it could be.  I suppose it depends on the person I'd work with.

 

Conclusion

This essay is good in making my career plans a lot less vague.  But it's full of guesses and I'm definitely missing a lot.  What do I plan on doing from here?

  • Get a significant amount of advice on this draft.  I'll be visiting England this summer and hope to stop by 80K and get advice from a bunch of cool people in person.  I'll sign up for another 80k coaching session (my first one got a bit derailed).  I also plan on talking to people not affiliated with 80K.  Just by passing this draft around, I hope to get more advice on how to modify it.  (Hint: You should give me advice on careers or advice on how to find good advice.)
  • Do more personal research.  80K has a lot of research, but they might not be going fast enough for me to make it in time, and I might need to pick up some of what they don't have, to the best of my ability.  Finding salary information for my favorite jobs would be a good start.
  • Choose a few paths to pursue and build toward them.  I can apply to jobs and graduate programs simultaneously while still figuring things out.
  • Work on the LSAT and GRE.  I need to take these two tests for law school and Ph.D. programs respectively.  I also might need graduate school for non-Ph.D. career advancement.  I'll get the practice tests ready.
  • Keep writing about this.  I'd imagine there's a lot of benefit for both me and others in creating more documentation and discussion around career choice for making the world a better place.
-

Also cross-posted on my blog.

New to LessWrong?

New Comment
58 comments, sorted by Click to highlight new comments since: Today at 6:52 PM

If you have enough uncertainty that you made this post, it seems like a good idea to take a gap year and spend the year researching more fully what your options are. The VoI is quite large. The main reason not to do this would be if you think taking a gap year would damage your chances of succeeding at at least one of your plausible options (e.g. it might look bad for grad school admissions); I don't know to what extent this is true though.

In particular, Peter might want to spend that year interning within some of his plausible career options.

This is in the works. Though, there still are enough potential careers that it would be impossible to intern for them all, so I would need to narrow things down.

This is a good point, but currently I think that I'll be able to gather sufficient information quickly enough to make a good initial decision. I can always decide on a gap year down the road if things don't work out.

if you think taking a gap year would damage your chances of succeeding at at least one of your plausible options

I'm am afraid of this.

Have you considered becoming an actuary? You need math, but it's a high grossing job (up there with finance, IT, and engineering) with long term prospects.

Also, in 10 months, you could probably pick up a fair amount for programming for an IT job.

Also, in 10 months, you could probably pick up a fair amount for programming for an IT job.

What would be the "next step" for pursuing an IT job if one is (1) generally "good at computers", (2) is capable of programming in PHP and JavaScript at an intermediate level, but (3) has taken only one actual class in computer science?

This all assumes you are decent at programming: It may still work even if you aren't, but become good while doing it.

  1. You put on your resume that you are x months away from receiving a Bachelor's Degree from Denison. No need to specify the major.
  2. You take some free online classes from Udacity or Coursera. At this time, you may become decent at programming, if you weren't already. You can now honestly say you've taken classes in computer science.
  3. Do so. No real reason to specify on your resume where you took the classes.
  4. List the technologies you know on your resume. Also, list some you could learn quickly, and technologies people associate with your specializations.
  5. Do some easy projects to gain some light practical experience in project implementation.
  6. Use this resume to pass the HR department's almost meaningless screening.
  7. Use your l33t skills to pass technical interview, and get job.
  8. ?
  9. Profit.

Should take a little less than a year to get to step 9.

I could imagine 7 having some difficulty. Could you elaborate or refer me to some references?

But this step-by-step guide actually seems quite reasonable.

I've heard this book is pretty good, but I've never read it. It's really easy to find copies of it floating around online:

http://www.amazon.com/books/dp/098478280X

It's still a field where demand outstrips the supply in places, so it may be possible to just land one by knowing enough to answer questions in the interview. I know currently certain bustling communities such as Drupal are dying to land people.

However, it's certainly better to have experience, and what you can do on that front is contribute to an open source project. It's incredibly easy to do nowadays, and valid as resume-worthy experience.

Have you considered becoming an actuary?

I hadn't; but I'll consider that now.

Have you considered becoming an actuary? You need math, but it's a high grossing job (up there with finance, IT, and engineering) with long term prospects.

The precise analysis of risk and uncertainty is also directly relevant to concerns in the LW/MIRI/GiveWell area.

On the other hand, the quip is that actuaries are people who found accountancy too exciting.

There are certain aspects of the salary/career choice issue which I almost never see discussed. For example, there's a tendency to fixate on starting salary and mean salary as the only metrics determining income. These are only two parameters of a salary distribution which needs several more parameters to fully characterize - the median, the standard deviation, and the skewness, and how these are all a function of time. Then there's what you might call the probability of promotion in a given year, which I'm fairly sure isn't a number you can look up online.

Then, after asking around and discovering that the probability of being promoted to the next salary grade in a given year is roughly 0.01, you have to ask yourself whether that number is low because it really is that hard and that competitive, or because most people are content to plateau at $100k/yr and/or lack the confidence to put themselves forward for promotion.

For example, in computer programming, engineering, and consulting, you can push your career down a management track toward becoming an executive within a large company. An executive position could command five or ten times the net income of a typical engineer. The frequentist odds of becoming an executive are very low, but I personally believe this is because almost nobody is actively trying to become one. Rather, most people who end up in executive positions are unusually ambitious, self-promoting, and somewhat lucky.

Investment Banking: My guess is this is the highest earning career, when considering probability of achieving certain incomes multiplied by the quantity of income, and high incomes right out of college.

Perhaps this is true, but I would challenge you over your use of the word "career." A career is a weird, organic, practically unplannable thing. If you browse the resumes of relatively successful people on LinkedIn, you will very rarely find anything that looks like

Worked at Company X for 25 years, performing the same general class of duties with gradually increasing responsibility.

Instead you will probably find a long list of companies, positions, roles, responsibilities, projects, and even titles - perhaps gradual transitions from "engineering" to "finance," or "computer programming" to "graphical design" to "marketing."

In fact, I would say such cross-pollination is almost a requirement! Nobody I personally know who works in banking started out in banking or finance or any of the "business" subdisciplines. Rather, they started their careers in other disciplines, became experts in the internal workings of those industries, and then leveraged that expertise to go into finance for those industries. Think about it from the perspective of the banks; do you want somebody with fifteen years of experience in finance only, or a guy with ten years of experience in engineering with five years of experience in finance as a natural progression of his increasing purview?

In summary, I feel that making a career choice is a nonlinear optimization problem. If it were easy to pick the highest-grossing career, then everybody would pick it, and it would cease to be the highest-grossing career. (This actually happens, repeatedly, as the new "hot thing" changes over time.) Investment banking seems too obviously the most lucrative choice, which indicates that in ten years it won't be anymore. There may be other factors you're missing. Maybe all those high salaries in banking are reserved for people with degrees from Yale, without which you might as well not bother. I don't know. The point is, career selection is full of these gotchas.

This is good advice; thanks.

Perhaps this is true, but I would challenge you over your use of the word "career." A career is a weird, organic, practically unplannable thing. If you browse the resumes of relatively successful people on LinkedIn, you will very rarely find anything that looks like "Worked at Company X for 25 years, performing the same general class of duties with gradually increasing responsibility."

True. Perhaps, I should clarify that I'm looking for my "first step". What job/career I would do first, before inevitably moving on to something else.

For example, there's a tendency to fixate on starting salary and mean salary as the only metrics determining income. These are only two parameters of a salary distribution which needs several more parameters to fully characterize - the median, the standard deviation, and the skewness, and how these are all a function of time.

Not to mention there's also a tendency to exclude benefits and bonuses from the calculation as well.

Not to mention there's also a tendency to exclude benefits and bonuses from the calculation as well.

PS: Also one must consider whether the employer has any donor matching programs, as that's basically additional salary as far as an "earner to give" is considered.

I recommend seeking out your field of comparative advantage, and then optimizing that. In other words, what area are you more likely to be more effective at than others? A large part of that is considering what you are likely to enjoy doing.

If you want to be a lawyer, then consider how you can be the most effective lawyer you can be; but otherwise stay away. There are many other people who will be better lawyers than you.

If you want to go into finance or investment banking, then do so. But don't do it just because it's the highest income opportunity. It's a tough field to break into, and you are unlikely to succeed without a lot of motivation. You may not succeed even with a lot of motivation, but if you are motivated at least you'll have fun trying.

If you want to go into software development, you're starting a little late; but you should spend your last 10 months in college getting as much of a head start as possible. In particular take your university's intro to programming class and whatever its successor is. At Denison it looks like that means 111, 173, 174, 271, and 371. Cram in as many of those as you can. If you like them and do well, that's a big clue that programming may be the most effective career choice for you. If not, cross it off your list.

Engineering you're unlikely to be successful at without an undergraduate engineering degree.

But overall, you're more likely to do more if you first decide what you want to do and then optimize to do that job most effectively, rather than picking a field because it seems likely to have a disproportionate impact.

In particular take your university's intro to programming class and whatever its successor is.

I've taken 111. Unfortunately, unless I go back to Denison for another semester, I'll be able to, at best, take one more computer science class.

Law: Law seems like a high-earning career if you can make it big, but there's a low chance of making it big. I've heard anecdotally that you basically need to get into a top law school or you won't be having a good time, assuming you can pay for that law school.

I endorse this perception.

I think this would be moderately enjoyable and with Moot Court, a Constitutional law class, and a Political Science major, I'm as prepared for law as anyone can be in a liberal arts school.

Practice at rigorous, high-quality writing (such as writing well-graded papers for class) is valuable for working a legal job. Political and moral theory, rather a lot less useful for the day-to-day, year-to-year work.

Recall that in law, you're mostly dealing with people who have a problem. Yes, there is some transactional work, but it is usually either low-paying or at least somewhat adversarial. Will dealing with miserable people all day make you miserable? If so, skip law.

For programming, if you follow the path that LW recommends, you should be able to tell within a month or two whether you like it enough to continue.

I would be surprised to learn that market research pays well. I just Googled, and this was the first hit that had salary data. It's not terrible, but any of the other careers (except law, if you're not in the top tier) will do better. That's not surprising; there are a ton of psychology and political science majors produced every year.

I would be surprised to learn that market research pays well.

I don't think it pays nearly as well as law / medicine / banking. I'd expect to enter at $50-60K. But I think I would enjoy it and think I could get into it easily, so it has those kinds of things going for it. And it doesn't require school beforehand (like law and medicine).

Out of the high-impact people you can think of, how many got there while not enjoying what they were doing and weren't that good at it? This should be enough of a clue that your first goal is to find something you both enjoy and have significant aptitude for. Of course, the best people often create their own careers (was there a meta-charity category before Give Well?).

was there a meta-charity category before Give Well?

Yes: Guidestar has been around for quite some time.

Of course, the best people often create their own careers

Probably not the best choice right out of college.

Or, at least, I don't presently have any good ideas.

Well, it's not like there's some future point at which you'll receive a certificate authorizing you to create your own career. On the contrary, it will always seem like a "weird" thing to do.

The younger you are, the less reason you have to be conservative and the better an idea it is to take risks that may result in large payoffs.

(It's actually not really a question of chronological age but rather of interpersonal obligations and embeddedness in social networks and relationships [having a spouse, children etc.], which tend to be lower in youth. But even [typically older] people who have such ties don't give enough consideration to doing "weird" things, in my opinion.)

Well, it's not like there's some future point at which you'll receive a certificate authorizing you to create your own career. On the contrary, it will always seem like a "weird" thing to do.

I'd want to pick up some experience and start-up income first.

A lot of the people who made their own careers (that I know of) didn't do it for their first career. GiveWell was made only after working for a hedge fund, etc.

I also have the "no good ideas" problem.

How long have you spent trying to solve the "no good ideas" problem?

It seems like the value of information here is ridiculously high. If you can think of even one good idea which enables you to start a career that you would love and that would allow you to make a huge impact on the world, certainly you could justify spending a whole year of your life just finding that one idea. But most people who make this statement can probably not report spending even a single hour applying their full effort to coming up with that idea.

I think most people who have a good career that they start based one one great idea went through dozens of good ideas in their mind before getting one great idea.

I don't think it's a good idea to try to convince someone who thinks he has no good ideas to go down that route if he's more comfortable taking an established path.

Entrepreneurs aren't made by being encouraged to think up an idea for a startup in an hour.

The problem is that there is a wide variety of careers which I can enjoy and of which (I think I) have aptitude for. How do I choose?

Also, the careers I (think I would) enjoy the most are not the careers I (think I) have the most aptitude for, and vice versa. How do I make that trade-off?

the careers I (think I would) enjoy the most are not the careers I (think I) have the most aptitude for, and vice versa

That narrows it down, keep looking until you find the intersection.

That's entirely unhelpful.

Aptitude is much less important than ability; if you enjoy the career, you will develop the ability soon enough.

This advice sounds like 'do what you're passionate about', which conflicts with the research done by 80k hours, though. See here and here.

These posts are well worth reading for anyone struggling with said advice, e.g. because they aren't passionate about anything. To paraphrase (though this doesn't do the 80k hours posts justice): Find something which is valuable and which you're likely to be good at, and you'll grow to enjoy it, too.

You didn't mention starting a business. You don't have any business ideas?

Correct. I have a very low expectation that I could start a business right out of college that would supply me with as much income as the other "earn to give" options.

[+][anonymous]11y-50