I want to open up the debate again whether to split donations or to concentrate them in one place.
One camp insists on donating all your money to a single charity with the highest current marginal effectiveness. The other camp claims that you should split donations for various reasons ranging from concerns like “if everyone thought like this” to “don’t put all your eggs in one basket.” My position is firmly in the second camp as it seems to me obvious that you should split your donations just as you split your investments, because of risk.
But it is not obvious at all. If a utility function is concave risk aversion arises completely naturally and with it all the associated theory of how to avoid unnecessary risk. Utilitarians however seem to consider it natural that the moral utility function is completely linear in the number of people or QALYs or any other measure of human well-being. Is there any theoretical reason risk-aversion can arise if a utility function is completely linear in the way described before?
In the same vein, there seems to be no theoretical reason for having time preference in a certain world. So if we agree that we should invest our donations and donate them later it seems like there is no reason to actually donate them at any time since at any such time we could follow the same reasoning and push the donation even further. Is the conlcusion then to either donate now or not at all? Or should the answer be way more complicated involving average and local economic growth and thus the impact of money donated now or later?
Let the perfect not be the enemy of the good, but this rabbit hole seems to go deeper and deeper.
Your utility function need not be completely linear, just locally linear. If your utility function measures against the total good done in the world, your effect on the world will be small enough to be locally linear
Most people don’t want to optimize the total good done, but instead care about the amount of good they do. People donate to charity until the marginal utility they derive from purchasing moral satisfaction falls below the marginal utility they derive from purchasing other things. In this case, diversification makes sense, because utility you assign to good you’re responsible for is very non-linear.
If you’re giving to charity at all, that’s awesome. Do what motivates you.
Interesting answer. Seeing as my personal giving is completely out of pleasure not some kind of moral obligation, the argument for diversification is very strong.
Diminishing returns. Giving $100M to a $10M problem because of information lag doesn’t actually allocate money nearly as efficiently as you might hope.
Do we have any evidence about the severity of that information lag? I can’t imagine it would be much more than a year.
I don’t know if this is a legitimatize concern. It may very well be worth a little waste value from info-lag for the benefit of overall efficiency.
Are we talking about lag within the community of people who read Givewell regularly, or in the broader community? Because that might be right for the former, but the latter has a lag that I expect to be measured in at least decades. Possibly centuries, given how much money still goes to churches.
I don’t think people who are giving their charity to churches are trying to give to the most efficient charity.
This is sort of the point. The more one is concerned with efficiency (like those who are concerned about the marginal benefit of a great charity vs. the best charity), the less an info-lag will have an effect on him/her. Effective Altruists who are thinking at the level of “I should only give to one absolutely best charity, to maximize marginal value”, are not likely to pick a charity and give to it blindly, never again looking at the least bit of evidence of marginal effectiveness.
I’m willing to wager that the marginal value lost due to an info lag when everyone is giving to only the best charity and aren’t paying enough attention, is vastly outweighed by the value produced due to everyone giving to the absolute best charities.
Sure, everyone should at a minimum aim to be very near the top, and the crowd who’d ever benefit from this discussion is the same crowd that’s least likely to suffer from lag. I’m attempting to make a theoretical case, not saying “Don’t donate to the best charities!”.
You get similar results by each individual giving to a different charity. As-is, there’s a few charities that are far above the rest, but if everyone thought like an effective altruist, that wouldn’t be the case. Consider the idea: if everyone bought the best TV then the factory that makes that model of TV would be inundated with orders.
That’s true but for now most of the problems are so large that even if almost everyone started thinking this way it would take a while before any high priority problem hit the point that it was no longer the best to donate to. Enough time for people to switch over to the next most important.
I believe donating to the best charity is essentially correct, for the reason you state. You won’t find much disagreement on Less Wrong or from GiveWell. Whether that’s obvious or not is a matter of opinion, I suppose. Note that in GiveWell’s latest top charity recommendations, they suggest splitting one’s donation among the very best charities for contingent reasons not having to do with risk aversion.
If you had some kind of donor advised fund that could grow to produce an arbitrarily large amount of good given enough time, that would present a conundrum. It would be exactly the same conundrum as the following puzzle: Suppose you can say a number, and get that much money; which number do you say? In practice, however, our choices are limited. The rule against perpetuities prevents you from donating long after your lifetime; and opportunities to do good with your money may dry up faster than your money grows. Holden Karnofsky has some somewhat more practical considerations.
So if we agree that we should invest our donations and donate them later it seems like there is no reason to actually donate them at any time since at any such time we could follow the same reasoning and push the donation even further. Is the conlcusion then to either donate now or not at all?
This depends on what someone believes the worthiest target for their donations is. If they’re trying to optimize for goals that require the continuance of Earth-originating intelligence, or humanity, than they’ll probably want to prevent human extinction. If they believe humanity faces a Great Filter or an existential risk in the coming decades, they’ll want to donate eventually. Supposedly they’d drive money to organizations will decrease the chances of humanity being destroyed. Of course, this still leaves the consideration that if they invest very well, or invest at the right time, (and know how to identify those things), such that money invested now for period t, increased to value n(x), will be worth more to the target organization at the end of period t, than the original value of money, x, is worth to the organization now.
However, its been argued from within effective altruism that the longer you wait to do good, the less the good you do will be worth. A donation is worth more now than later; effort applied at a later time will result in less value than the same level of effort applied now. This is called the haste consideration. Over time, the marginal increase in value of an investment over time comes up against the (presumed) marginal decrease of donation at any given time. I figure one could try measuring or calculating the rates of change here, and then figuring out some point at which the curves of them cross which would provide the optimum time to withdraw money from investment and donate it for maximum impact. However, that seems difficult. I’m not aware of something like this previously being done, and I follow effective altruism closely. So, if this has all been quantified, that report hasn’t been widely circulated.
This seems all the more important on LessWrong, because users on this site are more likely to care about existential risk reduction, whether they identify with effective altruism or not. Also, one could estimate a time past which investing rather than donating would be pointless, because that is beyond the point at which one expects humanity to preserve anything worth either investing in or donating to.
you should split your donations just as you split your investments, because of risk.
Isn’t it the case that most investment opportunities have essentially the same expected returns, due to market efficiency? In that case you want to diversify, since you can lower the variance without lowering the expected return. But if you can identify a single giving opportunity that has a significantly higher expected return than the alternatives, then it seems like you’d want to concentrate on that one opportunity.
Most people give to charity because it makes them feel good—knowing you’re helping people is a warm fuzzy feeling that most people enjoy. Obviously this can lead to irrationality pretty easily—look at the ineffective charities kept alive by nice narratives—but if we take that as the base reason, then standard human loss aversion can explain splitting. Your goal isn’t to improve the world per se, but instead to have your money improve the world. In other words, the argument about linearity of utility disappears, because one bad decision will destroy all the value you get from charity, and since that value is partially independent of the expected value of the good done in the world, this can happen even if you’re investing in the charity with the highest EV.
I don’t 100% agree with this, but it’s fairly close to my gut feeling—I split my political donations, but not my humanitarian donations.
I want to open up the debate again whether to split donations or to concentrate them in one place.
One camp insists on donating all your money to a single charity with the highest current marginal effectiveness. The other camp claims that you should split donations for various reasons ranging from concerns like “if everyone thought like this” to “don’t put all your eggs in one basket.” My position is firmly in the second camp as it seems to me obvious that you should split your donations just as you split your investments, because of risk.
But it is not obvious at all. If a utility function is concave risk aversion arises completely naturally and with it all the associated theory of how to avoid unnecessary risk. Utilitarians however seem to consider it natural that the moral utility function is completely linear in the number of people or QALYs or any other measure of human well-being. Is there any theoretical reason risk-aversion can arise if a utility function is completely linear in the way described before?
In the same vein, there seems to be no theoretical reason for having time preference in a certain world. So if we agree that we should invest our donations and donate them later it seems like there is no reason to actually donate them at any time since at any such time we could follow the same reasoning and push the donation even further. Is the conlcusion then to either donate now or not at all? Or should the answer be way more complicated involving average and local economic growth and thus the impact of money donated now or later?
Let the perfect not be the enemy of the good, but this rabbit hole seems to go deeper and deeper.
Your utility function need not be completely linear, just locally linear. If your utility function measures against the total good done in the world, your effect on the world will be small enough to be locally linear
Most people don’t want to optimize the total good done, but instead care about the amount of good they do. People donate to charity until the marginal utility they derive from purchasing moral satisfaction falls below the marginal utility they derive from purchasing other things. In this case, diversification makes sense, because utility you assign to good you’re responsible for is very non-linear.
If you’re giving to charity at all, that’s awesome. Do what motivates you.
Interesting answer. Seeing as my personal giving is completely out of pleasure not some kind of moral obligation, the argument for diversification is very strong.
Ah. Well, then there doesn’t seem to be anything to debate here. If you want to do what makes you happy, then do what makes you happy.
The theoretical question still stands.
...then, what? Every dollar of charity would be most efficiently invested and we’d solve the highest impact problems before moving on to others?
What happens if everyone thinks like this?
Diminishing returns. Giving $100M to a $10M problem because of information lag doesn’t actually allocate money nearly as efficiently as you might hope.
Do we have any evidence about the severity of that information lag? I can’t imagine it would be much more than a year. I don’t know if this is a legitimatize concern. It may very well be worth a little waste value from info-lag for the benefit of overall efficiency.
Are we talking about lag within the community of people who read Givewell regularly, or in the broader community? Because that might be right for the former, but the latter has a lag that I expect to be measured in at least decades. Possibly centuries, given how much money still goes to churches.
I don’t think people who are giving their charity to churches are trying to give to the most efficient charity.
This is sort of the point. The more one is concerned with efficiency (like those who are concerned about the marginal benefit of a great charity vs. the best charity), the less an info-lag will have an effect on him/her. Effective Altruists who are thinking at the level of “I should only give to one absolutely best charity, to maximize marginal value”, are not likely to pick a charity and give to it blindly, never again looking at the least bit of evidence of marginal effectiveness.
I’m willing to wager that the marginal value lost due to an info lag when everyone is giving to only the best charity and aren’t paying enough attention, is vastly outweighed by the value produced due to everyone giving to the absolute best charities.
Sure, everyone should at a minimum aim to be very near the top, and the crowd who’d ever benefit from this discussion is the same crowd that’s least likely to suffer from lag. I’m attempting to make a theoretical case, not saying “Don’t donate to the best charities!”.
You get similar results by each individual giving to a different charity. As-is, there’s a few charities that are far above the rest, but if everyone thought like an effective altruist, that wouldn’t be the case. Consider the idea: if everyone bought the best TV then the factory that makes that model of TV would be inundated with orders.
That’s true but for now most of the problems are so large that even if almost everyone started thinking this way it would take a while before any high priority problem hit the point that it was no longer the best to donate to. Enough time for people to switch over to the next most important.
I believe donating to the best charity is essentially correct, for the reason you state. You won’t find much disagreement on Less Wrong or from GiveWell. Whether that’s obvious or not is a matter of opinion, I suppose. Note that in GiveWell’s latest top charity recommendations, they suggest splitting one’s donation among the very best charities for contingent reasons not having to do with risk aversion.
If you had some kind of donor advised fund that could grow to produce an arbitrarily large amount of good given enough time, that would present a conundrum. It would be exactly the same conundrum as the following puzzle: Suppose you can say a number, and get that much money; which number do you say? In practice, however, our choices are limited. The rule against perpetuities prevents you from donating long after your lifetime; and opportunities to do good with your money may dry up faster than your money grows. Holden Karnofsky has some somewhat more practical considerations.
This depends on what someone believes the worthiest target for their donations is. If they’re trying to optimize for goals that require the continuance of Earth-originating intelligence, or humanity, than they’ll probably want to prevent human extinction. If they believe humanity faces a Great Filter or an existential risk in the coming decades, they’ll want to donate eventually. Supposedly they’d drive money to organizations will decrease the chances of humanity being destroyed. Of course, this still leaves the consideration that if they invest very well, or invest at the right time, (and know how to identify those things), such that money invested now for period t, increased to value n(x), will be worth more to the target organization at the end of period t, than the original value of money, x, is worth to the organization now.
However, its been argued from within effective altruism that the longer you wait to do good, the less the good you do will be worth. A donation is worth more now than later; effort applied at a later time will result in less value than the same level of effort applied now. This is called the haste consideration. Over time, the marginal increase in value of an investment over time comes up against the (presumed) marginal decrease of donation at any given time. I figure one could try measuring or calculating the rates of change here, and then figuring out some point at which the curves of them cross which would provide the optimum time to withdraw money from investment and donate it for maximum impact. However, that seems difficult. I’m not aware of something like this previously being done, and I follow effective altruism closely. So, if this has all been quantified, that report hasn’t been widely circulated.
This seems all the more important on LessWrong, because users on this site are more likely to care about existential risk reduction, whether they identify with effective altruism or not. Also, one could estimate a time past which investing rather than donating would be pointless, because that is beyond the point at which one expects humanity to preserve anything worth either investing in or donating to.
Isn’t it the case that most investment opportunities have essentially the same expected returns, due to market efficiency? In that case you want to diversify, since you can lower the variance without lowering the expected return. But if you can identify a single giving opportunity that has a significantly higher expected return than the alternatives, then it seems like you’d want to concentrate on that one opportunity.
Most people give to charity because it makes them feel good—knowing you’re helping people is a warm fuzzy feeling that most people enjoy. Obviously this can lead to irrationality pretty easily—look at the ineffective charities kept alive by nice narratives—but if we take that as the base reason, then standard human loss aversion can explain splitting. Your goal isn’t to improve the world per se, but instead to have your money improve the world. In other words, the argument about linearity of utility disappears, because one bad decision will destroy all the value you get from charity, and since that value is partially independent of the expected value of the good done in the world, this can happen even if you’re investing in the charity with the highest EV.
I don’t 100% agree with this, but it’s fairly close to my gut feeling—I split my political donations, but not my humanitarian donations.