I am writing this after significant difficulty making personal decisions in relation to career trajectory. For context I am a junior medical doctor, who particularly enjoys surgery. I'd come across the ideas of effective altruism about 18 months ago, and have seriously considered the pros and cons of a move to working on biosecurity.
The paths are quite different, with specialisation as a medical practitioner being a more concrete and well travelled path, whilst working on biosecurity seeming more nebulous and uncertain. After speaking with 80k, and with a number of people working in the biosecurity field to try and further characterise what a career may look like, I still am finding the decision difficult.
A significant part of the decision seems to be how much weight one should put on impact as part of the overall career decision equation. Should 'impact' be considered in a silo, in isolation from other aspects (such as compensation, satisfaction etc), or is its importance so much greater that it should be the main consideration.
These are a first pass at my thoughts on the topic, and would love to get feedback on how others think of this issue.
The notion of increasing one's reach and ability to do good is something that many strive for. Necessarily, it takes the individual away from the path that may have otherwise been forged, in search of a higher impact avenue. This can mean making quite significant diversions from what might have otherwise been an attractive path. It will also likely mean sacrificing the optimisation of other aspects of one's career in the pursuit.
Traditionally, there are a number of aspects to be considered when making a decision on which career path to pursue. These may include things such as compensation, the day to day satisfaction or fun of the work, altruistic motives, forging a sense of purpose, and the social status attached to the role. The aspects of a career choice may be more numerous, or divided up differently than this, but this is the distinction I will draw for the purposes of this post. Each person places a different weight on each aspect of any one career and decides what is the best fit for them to pursue.
This is an active process, and if done correctly means that the individual does not simply default into any random career path. They reflect, make weighted conscious decisions, act accordingly, and iterate the process to ensure they’re staying on track as time goes on. As their values change over time, they too may shift focus or direction.
The career advice within the effective altruism community argues for a much greater focus on the altruistic facet of career choice. Given the interconnected nature of the aspects of a traditional career choice framework, this tends to work relatively well. Compensation is seen as a means of personal resource acquisition, which would be best used establishing a level of personal comfort with any excess being used altruistically. Focus areas are often seen as meaningful, giving a strong sense of purpose and satisfaction. Many roles themselves are prestigious, and if not, the prestige is for-gone in place of more noble factors.
In this framework, the issue of fanaticism (that the expected value calculation will be dominated by a fractional credence in a choice with incredibly high stakes) dominating the set of factors is something that needs to be taken into account. This is of particular importance when considering working on existential risk. If there is a career path that would be less enjoyable and mean strain on personal relationships, but have a chance at moving the needle on a catastrophic global risk, should I take it? The expected value calculation to humanity would echo a resounding yes, but of course it is not that simple.
The mashing together of the aspects of career choice to a single measure of value is not something done deliberately. In fact there is express direction from EA organisations (80k) to consider aspects of a career outside of impact, such as personal fit, with a significant amount of weight. I believe collapsing to a singular focus on impact is more a fault of reasoning on a personal level. It is very difficult to maintain a balanced view of ones trajectory, and much more simple to focus on something (somewhat) quantifiable like impact.
The issue of the collapse to a singular focus on impact can put at risk the very essence of the project itself. Becoming narrowly focussed on something not directly tangible, such as ‘impact’ can be a recipe for burnout. Especially if it draws you away from things you may otherwise find intrinsically rewarding and enjoyable, but that may have a smaller amount of direct utility. This can be seen in a somewhat similar light to the more commonly seen issue of people seeking to maximize their income, to the detriment of any sense of enjoyment or work-life balance. It is almost never sustainable, and inevitably leads down paths that may have otherwise been seen as undesirable.
Focussing on impact is different from focussing on money in significant ways. The question is how to continue with the pursuit of a higher impact, without compromising on the equally important other, more ‘personal’ aspects of career choice.
I believe it is important to consider the various aspects of a career choice, namely compensation, satisfaction, impact, purpose, social standing, in a more discrete fashion as a first step. Each aspect should be considered independently prior to making trade-off between them. For instance, for a particular choice, say being a clinically practicing doctor, each of the aspects should be considered independently, then the same should be done for another option, such as a full-time academic working in biosecurity research. One should then consider how much weight they would want to give to each aspect. Perhaps it is 50% impact, with the remaining 50% divided evenly between compensation, satisfaction, purpose and social standing, perhaps it is some other division. This division needs to be honest, and as much as possible an accurate reflection of what the individual thinks to be true of themselves. Then, the choices can be weighed accordingly.
This approach avoids the potential for a single option to landslide knock-out alternatives because of a potential extreme result from one particular aspect. The process should not be a simple addition of the aspects, it should be a weighted sum. Further, we should iterate and constantly reflect, as we move forward and gain new information, we should update our decisions and readjust accordingly. At all stages this should be as active as possible, and we should aim to recalculate the weighted sums so as to not slip into a narrow focus. I believe this approach would allow one to follow a more personally satisfying, and overall more impactful career path.
I definitely agree that it's important to consider facets of a career other than impact, but personally I wouldn't want to use the weighted sum approach. I'd prefer to mostly think more about thresholds and aim to 'satisfice' in most areas other than impact.
It's very important to have boundaries and avoid burnout but if you have a reasonable sense of what level of social standing, income etc. is satisfactory and ensure that any career choice satisfies these conditions I think you can avoid these issues.
The reason I would prefer a satisficing approach for these aspects is, as you pointed out, that optimisation is costly, both in terms of cognitive resources and potentially in terms of happiness. Optimising a career along a number of dimensions is a very complex problem, as you pointed out, particularly when we have incomplete information and struggle to predict our future preferences. Furthermore, continually evaluating whether our career path is 'good enough' on a number of dimensions focusses our mind on the negative aspects of our career rather than being grateful for what we have.
And while it's incredibly important to ensure that you're appropriately tending to your wants/needs other than impact, impact has a broader range than the other factors which makes optimisation more important (one path may be orders of magnitude more impactful than another, but is less likely to differ to such a degree on other dimensions).
This approach might also be worthwhile as much of what these other factors are aiming for is likely to be something like happiness, and satisficing is more likely to be conducive to happiness (https://psycnet.apa.org/record/2002-18731-012).