You have 80,000 hours in your career. Here's how to do the most good with them.

Will MacAskill: Effective altruism is about trying to do the very most good that you can, making not just a difference but the most difference you can, and then taking an evidence-based, scientific approach. And then, finally, putting those ideas in practice, actually trying to do the most good in your life.

Will MacAskill:There are some outstandingly effective ways to donate. Among the more concrete and measurable ways to have an impact are GiveWell's top recommended charities. For example, by giving to the Against Malaria Foundation and distributing long-lasting and effective insecticide-treated bednets, you can save a life for $3,340. That's an actual statistic — you should expect to save a life. It's not the kind of marketing-speak you get from normal charities. There's also GiveDirectly, which gives direct cash transfers to the very poorest people in the world.

Different causes vary by quite a bit in terms of how much good you can do. Global poverty is exceptionally important, highly tractable, and still quite neglected among people hoping to do good. Factory farming is also in that category. Vast numbers of animals are used in factory farms and suffer absolutely needlessly. People of all sorts of different values would agree with that.

"I talk to a lot of people who are very disillusioned by working in the nonprofit sector, because you often feel like you're achieving nothing"

We also worry about small risks of global catastrophe where there's a very high expected cost. There are small probabilities of events with very large, very bad outcomes, with risks coming especially from new technology, like the development of synthetic biology and the potential for large-scale pandemics. Those are some of the cause areas effective altruists think are most important.

Will MacAskill:Many people should seriously consider "earning to give," which is deliberately taking a lucrative career in order to do good by your donations rather than deliberately working somewhere where you could have a direct, immediate impact. A number of people have gone on to do that as a career path.

Another very high-value career is working in politics. For some people, for example people coming out of Oxford's philosophy, politics, and economics (PPE) course, the chances of ultimately becoming a member of Parliament can be as good as 1 in 3, and chances of becoming prime minister of the United Kingdom can be as good as 1 in 300, so there's potential to have an absolutely huge impact.

We also recommend doing high-value research. Innovation in general is an extraordinary way to make a difference in the world. Entrepreneurship is another path we recommend.

Will MacAskill:Obviously there's some worry that you're disconnected and lose your values, but I'm coming around to the idea that the rate of doing that via earning to give is no worse than the rate of doing that through direct impact. I do talk to quite a lot of people who are very disillusioned having worked in the nonprofit sector or directly worked in poor countries, because you often feel like you're achieving nothing. It's the real-world implication of the theory, which we knew, that the majority of social programs have very little impact. That's very hard. You're doing this job that pays very poorly, is very demanding, and you're not convinced it's having an impact.

By contrast, if you're earning to give, you're in a cushy lifestyle — you're giving away 50 percent, but you're still on a nice salary — working with very smart people, and you know that the impact you're having is absolutely huge because you're able to donate to these very well-evidenced charities. I was worried about burnout and sustainability to begin with, and it's certainly still a concern I'll keep an eye out for. But it's become less pressing over time as people have actually been pursuing this path.

Will MacAskill:One is just, "Do you think you have unusual skills or interests in this area?" Matt Wage, who was profiled in the New York Times, was considering working at a nonprofit, was considering a philosophy PhD, but he actually just enjoyed quantitative trading the most. It's the sort of thing he could get up and work very solidly in a flow state for a long time. That's one important thing, because of sustainability and how well you'll do. If you think you can excel in a job, it's an extremely important factor for your impact as well.

"If you think you'd be a really good fit for politics, in particular, I'd really strongly encourage you to go into that"

The second thing is "How much do you think you can earn?" The difference in earnings between different professions is vast. Even within finance, the top quantitative trading firms can pay twice as much as even the top investment banking firms, and that makes a big difference to the potential impact you can have.

Then there's the question of what your other options are. If you think you'd be a really good fit for politics, in particular, I'd really strongly encourage you to go into that. That is something that can be extremely high-impact, but most people don't think they'd be able to stick at it. That makes it even more valuable — if you could be a good fit, if you're from a political family or something — to try to go into that field.

The final thing is that the case for earning to give does vary based on what cause you're concerned with. Some causes are more money-poor and some are more people-poor. Something like innovation and research, risks of global catastrophe, those causes seem to be more people-poor than money-poor. For them, it's more important to have really, really good people working at it. Animal welfare or global poverty, to some extent, there's more of an argument that the big problem is a lack of funding.

Will MacAskill:Entrepreneurship, especially tech entrepreneurship, ranks very highly among our recommended careers. It's one of our most recommended. There are a few reasons for that. One is that when you start out, you should weight career capital very highly. If you become an entrepreneur, you build up a wide variety of general-purpose skills. That's useful both for the skills and for what you learn about what you're good at yourself. Maybe you're forced to do coding and management and sales and design — all of these different things all at once — and that means you can figure out what you're good at. And it's a high-pressure situation, which is great for personal development.

Then there's the impact you can have through producing an innovative company, and that can be absolutely huge as well. Wave is my go-to example of this, co-founded by Lincoln Quirk, who's part of the effective altruism community. Its project is just to make remittances cheaper. Half a trillion in remittances are sent back to poor countries from migrants every year, over double total foreign aid flows. But at the moment, it's actually quite hard to do so. You have to go to a Western Union; the Western Union takes 10 percent. Wave is able to do it from phone to phone while taking only 3 percent. The potential increase in flows to poor countries is in the many billions of dollars per year, and they've already moved millions in remittances.

Will MacAskill:In terms of the impact you can make, one thing is to look at the odds for people in your reference class. The first person we advised on this was a white man coming from Oxford PPE, which is a very good reference class to be in if you want to be a politician in a country as elitist as the UK.

So if you're a white man in Oxford PPE, then what are your odds of becoming an MP, a Cabinet minister, or prime minister? The lower bound is to assume that all of your impact comes from becoming a successful party politician. The impact can be higher than that, because you can do great things in policy as well. You look at how much in the way of government spending you can swing, how much effect on legislation you can have once in office. Then you can just start to do the math in terms of how much of an expected impact you're making.

You mention the concern that maybe you wouldn't make much of a difference, and maybe you think that because it's so determined by the status quo and inertia. But those two considerations cancel out. If you think it's really hard to make a difference in politics, that means that when you do make a difference it has a bigger impact, because it stays for so much longer. All that consideration does is make the impact last longer.

"The PhD program we recommend most of all, especially if you're mathematically able, is economics"

There's a different worry, which is that you're not going to be able to do anything that the populace doesn't want, because it's a democracy, so you're not going to be able to do any actions that wouldn't already have been done by the person in your shoes. Having gotten to know quite a number of people in the UK political scene, it seems like this is clearly not the case. The clearest case to me is the foreign aid budget in the UK, which has been put at 0.7 percent. If you actually poll people, it's incredibly unpopular. But it seems that the ruling class understands how important foreign aid is and are happy to downplay that as a political issue and plow ahead with it.

There are potentially other examples as well. There are some issues that are so hot, like immigration, where it's very hard to do things because it's really in the mainstream media. But if you want to make a difference on things most people don't care about, like how we should handle synthetic biology, then it's not like the popular electorate has very strong views on that. That's something where you've got a lot more autonomy as a politician.

Will MacAskill:The PhD program we recommend most of all, especially if you're mathematically able, is economics. That's for a few reasons. One is because your potential impact is very great indeed as an economist. Because economists have such sway over policy, it's the subject where technocrats are having a really big influence, more so than any other. And there's huge amounts you can do. The revolution in development economics as a result of randomized controlled trials, led by Michael Kremer and Esther Duflo, only happened in the last couple of decades.

There's huge potential for much greater innovation in development economics, and in other areas as well. At the moment, macroeconomics in particular is not a science yet. We don't really understand it at all, we don't really know what's going on. That has huge implications, like financial crises and massive recessions that are incredibly damaging to the world. If we have more amazing people trying to figure that out, there's a clear path by which that has a massive positive impact on the world.

Then there's the second, more pragmatic aspect. If you get an economics PhD from a good university, and you want to become an academic economist, you probably will. The parity between the number of PhDs and the number of jobs is just much closer, and partly that's because so many people leave academia and work for the World Bank or in consulting or finance. There are a huge number of other things you can do with an economics PhD. That's again a further reason for a PhD in economics rather than other things: It's particularly flexible.

"One thing we don't recommend is working for a charity straight out of university"

In contrast, if you did a PhD in philosophy — I think there are a good number of extremely important research topics within academic philosophy. Even though many people would think of it as a joke, I think you can be a very high-impact philosophy researcher. But there are very few jobs available, and if you've got a philosophy PhD it's not very transferable. It's not useful for very many other things.

That makes the case for economics much stronger, and even stronger than things like a biology PhD. There's obviously amazing potential for innovations in biology. The contributions to human health have been one of the major success stories of the 20th century. But again, biology PhDs are far oversupplied relative to the number of jobs.

There are other areas of research that I think people may not have thought about as much, but are really important and slightly more abstract. One is good decision-making, and in particular group decision-making. There's a literature in psychology on just how bad people are at making decisions in groups. Yet that's how most decisions of great importance are made, whether on a board or by a group of employees or politicians on a committee. Developing ways by which people can make decisions better is an abstract way of making a difference, but it'd improve all elements of the world.

Will MacAskill:One thing we don't recommend is working for a charity straight out of university. Most charities don't have much impact; 75 percent of social programs, when tested, actually don't have a measurable impact at all. Some even are negative to society, like the Scared Straight program. And you don't get very good skills.

If you're aiming to do good with your career, you should be thinking, "How can I do a lot of good in the long run? In particular further on in my career, when I'm most influential?" That means that in the short run, when you're straight out of university, you should really be aiming to build up career capital, general-purpose skills, generally making yourself a better person so you can have more of an impact later on. Working in the charity sector isn't a great way to do that.

"Imagine if you smashed down the door to a burning building and rescued a child. You can do that every single year just by donating to these charities."

Will MacAskill:GiveWell's top recommended charities provide a baseline: Here's how much good you can do. If you're on only a little more than the typical income in the United States, just by giving 10 percent of your income you can save a life every single year.

Imagine if you smashed down the door to a burning building and rescued a child. That would stay with you for the rest of your life. You can do that every single year just by donating to these charities. It demonstrates that we can make an extraordinary difference.

The question is can we do even more through things that are harder to quantify, through organizations that pursue more systemic change? Very plausibly we can. So please, please do not conflate effective altruism just with evidence-based, RCT-backed charity.

If you think donations are obviously not the best thing you can do or obviously far inferior to other things you can do, I think that's naive in a different direction. I think that's failing to appreciate how uncertain and unreliable political change is. It's not always going to be good. We have a very poor understanding of large-scale changes to the world, and it's obviously also very hard to effect that sort of political change.

But some sorts of political change — campaigns for expanded immigration, to end the EU's common agricultural policy, better behavior from the International Monetary Fund, better trade deals that better serve the interests of the global poor — yes, I'm very in favor of that. There's then a difficult question about how you'd compare that, how you'd come to a decision about how the best things in systemic change compare to the best things with easier-to-measure interventions. I think it'll just require a lot of work from smart people working very honestly and transparently. That's starting to happen with the Open Philanthropy Project, where they're deliberately looking into policy change and trying to take a very honest, impartial, transparent approach to trying to make those comparisons between the Against Malaria Foundation and charities with a potentially much larger and harder to measure impact.

If people do want to donate to a group that works on systemic change, the one I'd recommend above all at the moment is the Center for Global Development, which is a think tank based in DC which does extraordinarily insightful analysis, and attempts to influence on global policy issues including immigration reform. Obviously it's very difficult to assess how big of an impact that's making, but it's certainly a very good use of money.