Sarah Lincoln

In this post, I will analyze the effectiveness of donating blood according to GiveWell’s top charity criteria.

Evidence of effectiveness

Red blood cell (RBC) transfusions are common practice in intensive care units; nearly half of patients receive one during their stay to increase oxygen delivery to tissues. This is unquestionably effective at increasing the survival rate of patients with profound anaemia, circulatory shock, or other conditions which result from critical impairment of tissue oxygenation.

That being said, those cases only make up a minority of blood transfusions. Most RBC transfusions are given to patients with otherwise high chances of survival, in the hopes of speeding up their recovery. The benefits of RBC transfusions for these patients are less documented, and it’s difficult to calculate how many QALY’s are actually saved.


To estimate the cost-effectiveness of blood donations, we can use the price of a unit of red blood cells in the UK - the equivalent of 156 US dollars - as a stand-in for its true value. Using this approach, red blood cell donations are 2 orders of magnitude less than an equivalent monetary donation to the Against Malaria Foundation in terms of cost-effectiveness.

However, many would argue that the cost of donating blood is not $156, it’s just 30 minutes of your time. The time-cost of that half hour is miniscule for most people, so blood donations could easily be on par with GiveWell’s top charities in terms of cost-effectiveness.

Room for more funding

As stated earlier, most blood transfusions go to patients who don’t necessarily need them for survival. Hospitals typically keep a store of O- red blood cell units for emergencies, and the existing group of dedicated blood donors generally produce enough donations to maintain these stores. Thus, additional donations from members of the EA community would likely go towards speeding up the recovery of patients with non-life-threatening conditions. While this is still a desirable outcome, some may argue that the marginal returns are too small to make donating blood a primary objective of effective altruism.

It’s important to note that there are some times when blood donations are more marginally effective, such as during the winter or holidays, when some core donors are unable to donate due to illness.

While the effectiveness of donating blood is almost certainly lower than donating money to GiveWell’s top charities, it comes at so little cost to the donor that many should at least consider it.


Guang Cui

Our Future

Guang Cui.jpg

Hey. My name is Guang. I’m just a college kid, typing in my dorm room—pondering our world, trying to decide what to pursue in my future. Having just completed a fellowship on effective altruism, I would like to share some of my thoughts on the movement, its ideas, and its direction.

Effective altruism is great. Growing up, I wondered why things were the way they were, feared a huge asteroid hitting us or a black hole swallowing us up, and tried to figure out how to make our home a better place. These are things we all speculate about, at least once in a while, and effective altruism gives a great framework for valuing these risks, opportunities, and ways of helping out.

As someone who loves math and understands its power, I appreciate EA’s focus on building up axiomatically and using numbers: we want to maximize some utility that measures human happiness, and use Bayes’ theorem and expected value to calculate benefits. The philosophy experiments about the drowning child and the fat man were super interesting to think about, and I believe there is value in looking past our emotional and evolutionary instincts to think “rationally”. I found Sam Harris’ talk to be a convincing argument for using science to solve moral issues. And I think EA is able to see the risk in artificial intelligence that most people overlook from analyzing things scientifically and mathematically: our brains are only a result of evolution (a slow process), and everything is made of an atomic structure that we can understand, so it’s very possible for us to simulate that artificially and produce something much smarter than we are.

There’s great insight in analyzing things quantitatively—in just “shutting up and multiplying”. I would say that it’s a way of thinking that everybody should learn.

However, I also believe that it’s not the only way we should think. EA is a young and potentially powerful movement, but there are plenty of people critical of it. Some arguments against EA seem ridiculous, but I think there are some valid points.

To me, “thinking mathematically” or what we typically term “rationally” is just one way of thinking. It helps us understand things better as human beings, but it’s not quite enough. We also need things like social and human skills and the ability to think artistically or historically. All of these are lenses through which we can view the world, and the more we have in our toolkit, the clearer our vision becomes. “Shutting up and multiplying” won’t solve everything.

There has been a stigma around rational thinking and effective altruism in that it’s ineffective or ridiculous, and that’s something we need focus on changing. Our numbers and theories aren’t perfect: there’s no “equality” factor built into our utility functions, and the perfect saint as measured by our standards doesn’t actually make for a model to aspire to. The introduction of “non-moral goods” means our theory is somewhat incomplete.
To expand effective altruism, we need to bring together people of more diverse backgrounds. We need people who work in arts, sales, and law, in addition to people in math, science, and philosophy. The good news is that our foundational movements are sound—we all want to improve the world—and we have a great framework and promising theories.

But the idea of improving the world is something that every human being on the planet thinks about, and should be able to give valuable input. Effective altruism should be a movement that encompasses all of us. By having diverse conversations and accepting the viewpoints of everyone, we will have a better sense of what constitutes a better world, and we can all work together to improve our future.

Imane Bouzit

Philanthropy vs. Time


Effective altruism is the use of evidence and reasoning to maximize the impacts of philanthropy. Leaders of this movement like Peter Singer, an Australian moral philosopher, believe that donating resources is imperative given the high levels of suffering--whether due to illness, oppression, or poverty--worldwide. While this ideal may seem obvious enough, numerous questions come to mind regarding the actual act of philanthropy: How much should be donated? To whom should resources be allocated, and according to what distribution guidelines? And when should donations be made? In this post, I will attempt to elaborate on the last question, although ultimately, this question will continue to inspire debate among effective altruists.

In a 2013 Effective Altruism Forum, writer Julia Wise provides a summary of arguments for either taking time before donating or donating immediately. With regard to waiting before donating, she explains that because effective altruism promotes quantitative analysis of service organizations, it takes time to determine the best intervention strategies. In this case, it makes more sense to give a particular organization time to improve its operations before supporting it financially. Likewise, investing in research might be a good way to expedite understanding of ideal interventions. Another argument Wise introduces is the possibility of investing now--whether in one’s education or career-- to have more wealth to donate later. I find this viewpoint to be attractive to the average effective altruist, as it encourages foresight in one’s career or business. Overall, the considerations of waiting for conclusive research about the best interventions or wealth accumulation are what I would categorize as the position of the first faction of effective altruists-- individuals who view extended time as being conducive to greater effectiveness.

The second main group of effective altruists are individuals who would argue for immediate intervention. Wise mentions that with time, some individuals may lose interest in donating as they age, investments may fail, and global problems which may have been mitigated sooner cause extended suffering. I naturally gravitate toward this second class of effective altruists because of the urgency of some causes-- the ongoing famine in Yemen or oppression of Uyghur Muslims being just two of numerous examples. Though I am a student with little to donate, I am willing to provide support. However, I am in favor of conducting research to find the most effective organizations for a certain cause as well as which organization would provide the most support per donation amount. Thus, although the debate of when to give may seem polarized to either immediate or culminating intervention, I posit that there is a third class of effective altruists: individuals who give within their capabilities while waiting for more effective intervention guidelines. While this approach has its flaws, it certainly addresses the need for urgency that underlies many global issues.

Adeline Hillier

The optimist in me would like to believe that Darwin was not fully correct when he proposed that our biological altruistic inclinations are reserved for the people who share our genes.  I’d also like to think that humans are not innately selfish, that there must be some way to account for the decision to act in the service of others outside our gene pool beyond the calculated benefits it brings to us.  One of the lesser-known facts about Darwin is his theory that humans wield an evolved tendency toward sympathy that supersedes the instinct for self-preservation, and that our resulting propensity for caretaking behavior and cooperation underlies the success of our species.  The “survival of the kindest” theory espouses that natural selection favors the occurrence of compassion rather than punishing it. 1

Most of us have heard that providing social support to others stimulates the primal reward center of the brain, releasing neurochemicals like oxytocin involved with the characteristic “warm glow” feeling of giving.  But there is further physiological evidence that humans may tend towards altruism by nature--for example, the vagus nerve may be adapted to promote altruism, and mirror neurons found uniquely in primates provide a possible neurological basis for the ability to empathize.2  But even if any sympathy we feel is not in fact instinctual, there still appears to be truth to the idea that social ties play an important role in prosocial behavior.  A 2005 study found that while non-Jews who had helped rescue people from the Nazis (as compared to others who did nothing) were less risk-averse, the strongest correlation was with those who reported having regular interactions with friends and family.3   A recent study suggests that altruistic behavior in children is not automatic, but grounded in a pre-established relationship with the person whom they choose to help. 4

I have introduced just some of many attempts to resolve the mystery of why people act altruistically at all--self-sacrificial behavior would seem at odds with our personal welfare.  There are likely many factors that play in. Psychoanalytic critics might support the theory that giving is merely a tool to reduce the burden on ourselves of our negative feelings about the situation, or of a possible guilt of not acting.  Some people might choose to act altruistically purely in the hope that the favor will be returned, or that it will appease a higher power and produce favorable future consequences for themselves, like getting into heaven or having a peaceful afterlife.  Or perhaps, sometimes, we can only be swayed by social pressures such as a desire to elevate others’ perceptions of us, a norm of reciprocity, or other desires or expectations that have taken root within a developed society that can make decisions that aren’t related to survival.  All of these things seem to count against the possibility that any intention is purely altruistic. One theory that I find particularly intriguing relates to an experiment in which students who had just spent time looking at a large tree or building were found to be significantly more likely than a control group to help retrieve pens that the experimenter ‘accidentally’ dropped on the ground (5).  A plausible explanation for this is that the self-loss that we experience in an awe-inspiring, or transcendent, moment reorients our thoughts in favor of prosocial behavior.  We see this effect in astronauts after they have witnessed the Earth from space--they return with a diminished sense of self and a newfound impulse to help others (the cognitive shift in awareness that occurs here is called the “Overview Effect”).  Whether some have a dispositional tendency toward such a state of mind, or if it has to be induced, perhaps we find reason to act altruistically by way of this kind of a third-party perspective in which we are removed from our own trivial interests and thus better able to think rationally.

If we are interested in mobilizing individuals for positive social change, perhaps it’s not enough to appeal to logic or unstated higher standards such as moral law.  Knowledge of an optimal way to act should be accompanied by a will to act. In the words of Matt Langdon, “the opposite of a hero isn’t a villain--it’s a bystander.”

1 Disalvo, D. (2009). Forget Survival of the Fittest: It is Kindness that Counts. Scientific American, 20(5), 18-19.

2 Keltner, D. (2012). The Compassionate Species. UC Berkeley Greater Good Science Center.

3 Midlarsky E, Fagin Jones S, Corley RP. (2005). Personality correlates of heroic rescue during the holocaust. J. Pers., 73(4), 907-934.

4 Carey, B.. (2014). Stanford psychologists show that altruism is not simply innate. Stanford News.

5 Piff, P. K., Dietze, P., Feinberg, M., Stancato, D. M., & Keltner, D. (2015). Awe, the small self, and prosocial behavior. Journal of Personality and Social Psychology, 108(6), 883-889.

Shulammite Lim

Biosecurity: not just a cold

On its list of most urgent global issues, 80,000 hours includes biosecurity, pointing to the millions—even billions—of people that could potentially be affected by natural pandemics and new, scientifically engineered pathogens. Biorisk reduction reaches a whopping 15 out of 16 in the scale category, as outbreaks of pathogens can cause global catastrophe. Thankfully, this cause has gained some traction, as it ranks a 4 out of 12 on the neglectedness scale based on the over $1 billion spent by the US government for biosecurity purposes. Solvability has been assessed at a 4 out of 8, as there exist some possible ways to make progress toward biorisk reduction, with some support from experts.

Biosecurity includes many disease-related risks, including natural pandemics, bioterrorism, and accidental deployment of biological agents in research. These can be differentiated from global health issues in that they have relatively low-probability risks and potentially global impact, instead of continuing health issues that may be smaller in scope. The Open Philanthropy Project estimates that natural pandemics are likely the largest current threat, but novel biotechnology could contribute to heightened risk in future years.

The worst flu pandemic in the 20th century was the Spanish flu epidemic of 1918, which is believed to have caused 50-100 million deaths, or 3-5% of the world’s population then. If a similar pandemic were to break out today, that same percentage of the population would equal 288,000,000 people. This could be explained with the pathogen being more likely spread quickly courtesy of international travel, though modern medical advances may also reduce their potential impact.

Bioterrorism looms as a fear-inducing prospect and could take many forms. Some possibilities include a noncontagious biological agent like anthrax (e.g. Amerithrax, aka 2001 anthrax attacks), a noncontagious natural pathogen that has been “eradicated” and is no longer vaccinated against (e.g. smallpox, with the exception of its presence in research labs in Atlanta, Georgia, and Russia), and a contagious engineered pathogen, which may be created in dual use research.

“Dual use” research refers to research that could be used for both positive and negative effects. For example, there has been much controversy brewing over research intended to alter the H5N1 flu virus to make it transmissible between ferrets, which are used as a model for humans. With this type of research, two fearful possibilities lurk: scientists working on this project may accidentally release a harmful agent, or they may create a pathogen that allows the ill-intentioned to do their will with greater ease.

Biosecurity has mobilized multiple entities to surveil emerging threats and step up on research on novel therapeutics. It is a focus area of the Open Philanthropy Project, who has noted that though the U.S. has directed much funding to various aspects of biosecurity since 2012, philanthropic involvement in this field is still rather limited. To take action, new philanthropists can make the most impact in research and advocacy. Possible avenues of action include reaching out to policymakers to improve biosecurity initiatives, supporting research in the field, promoting stronger oversight of dual use research, and growing stockpiles of crucial medical countermeasures.

Swapnil Garg

Existential Risk: Good or Bad?

In Effective Altruism, a central question is that of what the biggest priority should be. Nick Bostrom is a well-known advocate for decreasing existential risk. In his paper “Astronomical Waste: The Opportunity Cost of Delayed Technological Development,” he notes a possible 10^23 possible human lives living at once in the Virgo supercluster. Thus, decreasing the probability of existential risk, or increasing the probability of eventual colonization of the universe, by even a very small amount could massively increase total utility. However, Phil Torres argues that space colonization will cause suffering on an astronomical scale, and therefore contends that it is most desirable to delay space colonization or prevent it from happening ( rule-have-catastrophic-consequences-i). It seems that either decreasing or increasing existential risk will cause the most good/decrease the most suffering for humans, so one of them should be the cause area with the highest priority. But which one? I argue that the highest priority should be given to determining whether decreasing or increasing existential risk is better.

I believe that astronomical suffering is much worse than astronomical colonization can be good. For one thing, extreme pain can be inflicted on the human body, while I don’t think the same thing can be said about extreme pleasure. In addition, Bostrom notes that some utilitarians may adopt a person-affecting view, where they primarily try to maximize the utility of past and ongoing human lives, and a loss of potential future lives is not bad. However, while they might not care about an astronomical number of future human lives, surely these utilitarians would be against those lives being in almost constant suffering. So, even if there is a very small probability of Phil Torres’s prediction of astronomical suffering, the expected amount of suffering is still very large, while the benefits from space colonization might not be as big. Therefore, more research on “suffering risk” should be the highest priority cause area.

Phil Torres also notes that there have been many individuals throughout history with the intention to cause as much suffering as possible. It seems almost certain that many more of these individuals will exist in the future if humanity doesn’t undergo extinction. With enough intelligence, drive, and access to large-scale weapons, they could conceivably cause astronomical suffering. If the proportion of such humans remains constant, then in the future, there would be large numbers of people with the intention to cause massive suffering, and if even one obtained access to the most advanced weapons, catastrophic suffering would result. So, more research into methods of decreasing such tendencies in people is also of the utmost priority.

Naomi Michael

Effective Altruism Blog Post: Criticisms of EA

For the past eight weeks, we have examined the motivations, doctrine, and application of the growing Effective Altruism movement. But, to more effectively tackle the world’s biggest problems, as EA hopes to do, we must also look at where we could fall short of this goal. Most criticisms of EA emphasize the faults in the “thick” rather than “thin” version of EA. The “thin” version is that people should do the most good they can. To disagree with this fundamental principle would be to reject EA altogether. But, critics more often take aim at the “thick” version of EA that consists of a wide range of assumptions and ideas associated with EA. Broadly, these ideas are welfarism, consequentialism, and scientific evidence. Speaking to welfarism, critics of EA often point to the fact that EA doesn’t give enough weight to equality or rights when deciding which causes to dedicate resources to. EA looks instead to maximize rather than distribute overall good. In regards to the other associated ideas, critiques typically center around the fact that EA can’t focus on systemic causes and structural inequalities when analyzing RTC or cost effectiveness experiments.   

The main takeaway from these critiques is that effective altruism may not by nature focus on the systemic causes of problems and may rather seek to mitigate immediate causes of harm and death. By focusing on these problems, however, EA is able to mobilize resources from the wealthiest parts of the world and allocate them to those in most desperate need. As the movement grows, space may emerge to broaden the scope of EA’s efforts and metrics to encompass sociopolitical factors as well as the currently used metrics of QALYS and RCTs. But, as of now, the goal of EA is to do charity in the most effective manner under the methodical assumptions that it is best to save the most lives possible, seek tractability, help where there is the greatest potential for increased funds, and do so impartially. Perhaps charitable efforts that can’t be measured with RCTs and hard data will be undervalued by the movement, but that doesn’t undercut EA’s ability to mobilize those with the time, talent, or financial resources to do immense amounts of good, particularly in regards to extreme poverty, existential risk, and global health. And at the end of the day, if your goal is to do the most good you are capable of, regardless of which career path you undertake, the framework of EA supports such an endeavor.