We wouldn’t pay much more to save 200,000 birds than we would to save 2,000 birds. Our willingness to pay does not scale with the size of potential impact. Instead of making decisions with first-grade math, we imagine a single drowning bird and then give money based on the strength of our emotional response to that imagined scenario. (Scope insensitivity, affect heuristic.)
People’s willingness to pay depends mostly on their income. I don’t understand why this is crazy.
UPDATED: Having read Nectanebo’s reply, I am revising my original comment. I think if you have a lot of wasteful spending, then it does make you “crazy” if your amount is uncorrelated with the number of birds. On hearing, “Okay, it’s really 200,000 birds,” you should be willing to stop buying lattes and make coffee at home. (I’m making an assumption about values.) Eat out less. Etc. But if you have already done these things, then I don’t see why your first number should change (at least if we’re still talking about birds).
Not all of a person’s money goes into one charity. A person can spend their money on many different things, and can choose how much to spend on each different thing. Think of willingness to pay to actually be a measure of how much you care. Basically, the bird situation is crazy because humans barely if at all feel a difference in terms of how much they give a damn between something that has one positive effect, and something that has 100x that positive effect!
To Luke: This person was reading about the biases you breifly outlined, and he ended up confused by one of the examples. While the linking helps a good deal, I think your overview of those biases may have been a little too brief, and they might not really hit home with readers of your site, and personally I think it might be difficult particularly for those who may not be familiar with the topics and content of the site. I don’t think it would be a bad idea to expand on each of them just a little bit more.
The point of the excerpt you quote has nothing to do with income at all; the point is that (for example) if I have $100 budgeted for charity work, and I’m willing to spend $50 of that to save 2,000 birds, then I ought to be willing to spend $75 of that to save 10,000 birds, because 2000⁄50 > 10000⁄75. But in fact many people are not.
Of course, the original point depends on the assumption that the value of N birds scales at least somewhat linearly. If I’ve concluded that 2000 is an optimal breeding population and I’m building an arcology to save animals from an impending environmental collapse, I might well be willing to spend a lot to save 2,000 birds and not much more to save 20,000 for entirely sound reasons.
If I budgeted $100 for charity work and I decided saving birds was the best use of my money then I would just give the whole hundred. If I later hear more birds need saving, I will feel bad. But I won’t give more.
Suppose you budgeted $100 for charity, and then found out that all charities were useless—they just spent the money on cars for kleptocrats. Would you still donate the money to charity?
Probably not—because hearing that charity is less effective than you had thought reduces the amount you spend on it. Equally, hearing it is more effective should increase the amount you spend on it.
This principle is refered to as the Law of Equi-Marginal Returns.
Yes, if saving birds is the best use of your entire charity budget, then you should give the whole $100 to save birds. Agreed. And, yes, if you’ve spent your entire charity budget on charity, then you don’t give more. Agreed.
I can’t tell whether you’re under the impression that either of those points are somehow responsive to my point (or to the original article), or whether you’re not trying to be responsive.
I was describing how I would respond in that situation. The amount I would give to charity XYZ is completely determined by my income. I need you to explain to me why this is wrong.
The amount I give to charity XYZ ought not be completely determined by my income. For example, if charity XYZ sets fire to all money donated to it, that fact also ought to figure into my decision of how much to donate to XYZ.
What ought to be determined by my income is my overall charity budget. Which charities I spend that budget on should be determined by properties of the charities themselves: specifically, by what they will accomplish with the money I donate to them.
For example, if charities XYZ and ABC both save birds, and I’m willing to spend $100 on saving birds, I still have to decide whether to donate that $100 to XYZ or ABC or some combination. One way to do this is to ask how many birds that $100 will save in each case… for example, if XYZ can save 10 birds with my $100, and ABC can save 100 birds, I should prefer to donate the money to ABC, since I save more birds that way.
Similarly, if it turns out that ABC can save 100 birds with $50, but can’t save a 101st bird no matter how much money I donate to ABC, I should prefer to donate only $50 to ABC.
Once upon a time, three groups of subjects were asked how much they would pay to save 2000 / 20000 / 200000 migrating birds from drowning in uncovered oil ponds. The groups respectively answered $80, $78, and $88 [1]. This is scope insensitivity or scope neglect: the number of birds saved—the scope of the altruistic action—had little effect on willingness to pay.
Now I haven’t read the paper, but this implies there is only one charity doing the asking. First they ask how much you would give to save 2000 birds? You say, “$100.” Then they ask you the same thing again, just changing the number. You still say, “$100. It’s all I have.” So what’s wrong with that?
Agreed: if I assume that there’s a hard upper limit being externally imposed on those answers (e.g., that I only have $80, $78, and $88 to spend in the first place, and that even the least valuable of the three choices is worth more to me than everything I have to spend) then those answers don’t demonstrate interesting scope insensitivity.
There’s nothing wrong with that conclusion, given those assumptions.
RE: The Crazy Robot’s Rebellion
People’s willingness to pay depends mostly on their income. I don’t understand why this is crazy.
UPDATED: Having read Nectanebo’s reply, I am revising my original comment. I think if you have a lot of wasteful spending, then it does make you “crazy” if your amount is uncorrelated with the number of birds. On hearing, “Okay, it’s really 200,000 birds,” you should be willing to stop buying lattes and make coffee at home. (I’m making an assumption about values.) Eat out less. Etc. But if you have already done these things, then I don’t see why your first number should change (at least if we’re still talking about birds).
Not all of a person’s money goes into one charity. A person can spend their money on many different things, and can choose how much to spend on each different thing. Think of willingness to pay to actually be a measure of how much you care. Basically, the bird situation is crazy because humans barely if at all feel a difference in terms of how much they give a damn between something that has one positive effect, and something that has 100x that positive effect!
To Luke: This person was reading about the biases you breifly outlined, and he ended up confused by one of the examples. While the linking helps a good deal, I think your overview of those biases may have been a little too brief, and they might not really hit home with readers of your site, and personally I think it might be difficult particularly for those who may not be familiar with the topics and content of the site. I don’t think it would be a bad idea to expand on each of them just a little bit more.
I suppose if you are the sort of person who has a lot of “waste”.
Edit: misread
This comment confuses me.
The point of the excerpt you quote has nothing to do with income at all; the point is that (for example) if I have $100 budgeted for charity work, and I’m willing to spend $50 of that to save 2,000 birds, then I ought to be willing to spend $75 of that to save 10,000 birds, because 2000⁄50 > 10000⁄75. But in fact many people are not.
Of course, the original point depends on the assumption that the value of N birds scales at least somewhat linearly. If I’ve concluded that 2000 is an optimal breeding population and I’m building an arcology to save animals from an impending environmental collapse, I might well be willing to spend a lot to save 2,000 birds and not much more to save 20,000 for entirely sound reasons.
If I budgeted $100 for charity work and I decided saving birds was the best use of my money then I would just give the whole hundred. If I later hear more birds need saving, I will feel bad. But I won’t give more.
Suppose you budgeted $100 for charity, and then found out that all charities were useless—they just spent the money on cars for kleptocrats. Would you still donate the money to charity?
Probably not—because hearing that charity is less effective than you had thought reduces the amount you spend on it. Equally, hearing it is more effective should increase the amount you spend on it.
This principle is refered to as the Law of Equi-Marginal Returns.
Yes, if saving birds is the best use of your entire charity budget, then you should give the whole $100 to save birds. Agreed.
And, yes, if you’ve spent your entire charity budget on charity, then you don’t give more. Agreed.
I can’t tell whether you’re under the impression that either of those points are somehow responsive to my point (or to the original article), or whether you’re not trying to be responsive.
I was describing how I would respond in that situation. The amount I would give to charity XYZ is completely determined by my income. I need you to explain to me why this is wrong.
OK, if you insist.
The amount I give to charity XYZ ought not be completely determined by my income. For example, if charity XYZ sets fire to all money donated to it, that fact also ought to figure into my decision of how much to donate to XYZ.
What ought to be determined by my income is my overall charity budget. Which charities I spend that budget on should be determined by properties of the charities themselves: specifically, by what they will accomplish with the money I donate to them.
For example, if charities XYZ and ABC both save birds, and I’m willing to spend $100 on saving birds, I still have to decide whether to donate that $100 to XYZ or ABC or some combination. One way to do this is to ask how many birds that $100 will save in each case… for example, if XYZ can save 10 birds with my $100, and ABC can save 100 birds, I should prefer to donate the money to ABC, since I save more birds that way.
Similarly, if it turns out that ABC can save 100 birds with $50, but can’t save a 101st bird no matter how much money I donate to ABC, I should prefer to donate only $50 to ABC.
From Scope Insensitivity:
Now I haven’t read the paper, but this implies there is only one charity doing the asking. First they ask how much you would give to save 2000 birds? You say, “$100.” Then they ask you the same thing again, just changing the number. You still say, “$100. It’s all I have.” So what’s wrong with that?
Agreed: if I assume that there’s a hard upper limit being externally imposed on those answers (e.g., that I only have $80, $78, and $88 to spend in the first place, and that even the least valuable of the three choices is worth more to me than everything I have to spend) then those answers don’t demonstrate interesting scope insensitivity.
There’s nothing wrong with that conclusion, given those assumptions.
Have you read Scope Insensitivty? It’s not just an income effect—human beings are really bad at judging effect sizes.
Of course, I’ve read it. My problem isn’t with scope insensitivity. Just this example.