I can think of one big advantage of charity over politics: Giving to politics is rent-seeking behavior. If you give money to support politicians against/for dairy tariffs, then you encourage the other side to give money to their politicians who are for/against dairy tariffs. In this game matrix, “cooperate” is don’t give money.
I have a hard time swallowing the ten trillion dollar figure. I don’t think the two parties are really that different, and there’s good, game theory for them to be that way.
I think there is a market for some sort of organization by which a person who wants to donate $50 to the Democrats can contact a person who wants to donate $50 to the Republicans and mutually agree to donate their collective $100 to some third-party cause like world hunger instead.
I like this idea, but how do you verify that the donation would have gone to the political party? If I’m inclined to give $50 to a non-political-party charity, then there’s incentive for me to claim it would go to the party I prefer in order to get one of my political opponents to divert money from the Bad Guys to another cause.
I would like to come up with aways to prevent gaming this sort of system. If we could get past this sort of hurdle, I would be interested in implementing a website facilitating this.
Perhaps the matched donors have to agree on a charity and an amount: you pay into the system, and if you get matched, then the non-political charity gets your money, and if you don’t (after some waiting period), then the political charity does. This means you have to at least not mind the money going to the political charity, or be willing to gamble.
A thought: Charity popularity probably isn’t evenly distributed between political parties. This could limit the ability of some charities to be matched, but it would allow (for example) a Republican who supports Planned Parenthood to have little or no risk of their donation defaulting to the GOP. A Republican who didn’t want to donate to the Republican Party and did want to donate to Planned Parenthood would take very little risk, while one who wanted to donate to the NRA might have more trouble.
Charity itself isn’t evenly distributed: Republicans give more dollars to charity, period. If everyone in the US signed up for this site, it would guarantee a Republican win. But, of course, not everyone will, so it may not be that bad.
Yes, that should reduce the gaming of the system. It involves more secrecy that in my original conception of the website (which would display graphs of the matching money available for different charities), but I don’t think that can be helped.
how do you verify that the donation would have gone to the political party?
It is probably impossible to become extremely confident about that, but past contributions to a party are a good predictor of future contributions, and I believe that Federal campaigns are required to disclose the names and amounts of any contributions to the Federal Election Commission, which publishes them.
Those published disclosures can be used to qualify a counterparty (a person one is considering making a deal with) and then to verify that the counterparty carried out his end of the deal.
The deal should be that a Republican promises to contribute X dollars less this election cycle than he did last cycle if a Democrat will do the same.
The deal reduces (direct) contributions to campaigns even without the requirement that the Republican and the Democrat agree on a charity to donate 2 * X dollars to. In particular, it reduces (direct) contributions even if the two “sides” of the deal just keep that money.
There are indirect ways to make contributions that do not require disclosure, however, and IIUC these indirect ways are heavily used because they allow individuals to get around the dollar limits on direct contributions. So that consideration brings back the idea that the $X each “side” saves in the deal should go to charity (but I see no need to require that each “side” donate their $X to the same charity, just that the donations can be verified) because it gives some assurance that the $X will not become indirect contributions to a political party, since the total amount an individual is willing to spend on altruism tend to stay relatively constant year over year. But that takes out of the stream of money going to campaigns and parties mostly altruistic money, leaving mostly unaffected the money that expects to profit from the contributions, which might have a bad effect on the political process.
Also, credible arguments have been made that past experience with the dollar-limit laws shows that it is futile to keep money out of politics because the people who make their living from that money have so much influence on the law-making process and because there is so much smart money that wants to contribute to candidates and to parties. Carl’s analysis in the OP of the efficacy of “buying” a vote is evidence for that last point.
The problem is that if the potential participants expect this type of cheating, that their donations will be diverted by someone who would not have donated to their political opponents anyways, they will not want to participate.
I think you have to just eat this problem. The whole point of the system is that it allows a donation to do double duty as a political force and as a standard charitable donation. It will appeal to anyone who likes both, whichever one is more important to them.
Maybe this should be handled at a different level: the two parties get together and agree to each take a large amount of their donation funds and redirect it to a third-party charity. This seems less game-able.
If the parties could get on board with this, they could provide a separate donation option for something like this. Both parties would then commit to give an amount worth the smaller of the sums gathered into charity. Now for the funders not to donate to the charity option would mean a smaller charity sum for their party and more campaign funds for the opposing party if their party ended up with the smaller charity sum.
For one, it eliminates the problem Alicorn suggests: we already know the money would go to the respective party because it has been donated already. Also, it would be easier to manage, because it would be a single public trade with a large amount of money involved, not many private $100 agreements.
I was thinking that a charity might propose this to each party individually and get them to agree to it. But upon reflection, a better way would be to run a charity (or meta-charity) that works like this: when you donate money, you specify your affiliation, and the charity takes care of donating the difference to the appropriate party.
Perhaps it’s not even necessary to specify a single neutral charity option. You give money to the meta-charity and specify your party affiliation and preferred charity. Then we donate the difference to the party’s campaign fund as usual, and split up the remaining money between charities in proportion to how much was meant to be donated to each.
Wow, I expect this kind of naivete from normal people, but not from LWers. This is exactly the sort of bias-influenced human behavior that LW should be teaching you to understand. It’s more Hansonian than Yudkowskian, but still.
Politics is not about policy. Donors are signaling affiliation.
Probably true, but that isn’t how donors see themselves. The idea here is to let them achieve the goals they proclaim, while still signaling their affiliation.
The idea here is to let them achieve the goals they proclaim, while still signaling their affiliation.
It would seem to be aimed at those with multiple signals they wish to send. (Political affiliation, charitable contribution, a general attitude along the lines of efficiency). It seems most likely to appeal to those who would otherwise be directly donating to charities but want a bit extra out of their contribution.
What if you give every donor a profile page that shows what causes you’ve defunded? What if you linked your facebook to a chart showing how much you’ve kept away from pro-life/pro-choice lobbyists?
Correct me if I’m wrong, but this seems like an opportunity to signal tons of group affiliation, intelligence, and compassion.
If money is diverted from small political donations by lots of individuals, does that increase the influence of big corporate donors over politicians?
Could this effect be ammeliorated by getting politicians themselves to value the cancelling of a donation to their opponents as much as a donation to themselves?
Corporations reap some PR effects from donating to charity, so in theory, they’d be even more motivated than individuals to accept diverted matched donations.
Trying to get people who make small donation to cancel out each other by donating to a third course has the opposite effect of creating public financing of elections.
Detail: the website takes donations in rounds that end on a specified date. Each donation has three parts: a political recipient (say, the DSCC), an offset recipient (say, the NRSC), and a backup charity (say, Village Reach). The backup charity must be scrupulously non-political. At the end of each round, the website finds all the donations which match, gives the political recipient who gets more money the difference in donations, and parcels out each side’s remaining money (which will be equal) proportionately among their named charities.
A $100 DSCC/NRSC/VillageReach donation is thus guaranteed to cause the difference in donations between the DSCC and NRSC to go $100 in the DSCC’s favour; I just don’t know whether it will do so by increasing the money the DSCC gets, or reducing the money the NRSC gets. But either way, if there are donors on both sides then VillageReach will get at least some of the money, and if they’re roughly evenly matched it’ll get most of it.
So the big question is: how could this be made to seem attractive to people? It seems like very few would understand it. “No, give to the DSCC directly - that way they get all the money! Otherwise some Republican will put money in and take away your donation!”
I think this is exactly how to do this. I’m not sure how effective it could actually be, but I’d really like someone to make an honest go for it.
One note: I think there is a motivation for underdog supporters not to contribute, since presumably the less-well funded candidate needs funding more than the better funded candidate. I think this should be relatively small though, at least for races that are at all contentious.
In theory you could state with each donation how much you’d need to take off the other side in order not to give it to your side and look for matches in a process a little like betting markets, but it’s hard to imagine that working out in practice, so maybe you just aim this tool at evenly matched races.
But politics is not about policy. Political donors want to signal their affiliation with their tribes, not spend their money efficiently to change the world. It’s a modern potlatch. Otherwise they’d be giving to something else in the first place—it’s extraordinarily unlikely that giving to the Democrats or Republicans is the maximal way to impact the world, so anyone who is doing it obviously doesn’t have the goal of efficient charity.
I predict that such an organization would capture < 0.5% of donors by # and < 0.1% by $. That’s a pretty small market—a few hundred thousand.
I wonder which type of voter is more likely to think such a system is a good idea. I would be surprised if both sides were equally open minded regarding such things. (But aren’t American so don’t know which lot is least insane.)
Such a system would appeal to the type of voter who doesn’t vote because he is rationally ignorant and/or calculates it isn’t worth his time (b/c he is not a utilitarian and doesn’t count common benefit as a reason to vote), and wouldn’t donate to a political party because he knows there are far more efficient ways to transform money into changing the expected future of the world. He can see why the system is an efficiency gain using the same tools he can see why voting and donating to political parties is a waste of his time and money.
This creates somewhat of a problem for the proposal, if it is only appreciated by those who can’t benefit from it...
Yes, I think this is a great line of thought (despite Alicorn’s legitimate objection). In light of the fact that there’s a fair amount of overlap in human values even among members of disparate groups, it seems like there’s a substantial arbitrage opportunity attached to redirecting partisan efforts to efforts that are unambiguously good.
Re the Prisoner’s Dilemma element in political competition, for this post I left it out (save the acknowledgement with respect to moral objections, and associated link to the Ezra Klein comments section). More on that to come.
Re the second point, the ten trillion dollar figure is not mine (my estimates and data are coming in subsequent posts). I have an upcoming post on the data regarding the applicability of median voter theorem style analysis in the U.S. (short answer: democratic competition constrains differences between candidates to a small fraction of the variation in outcomes with dictators, but different politicians representing the same electorate thanks to narrow wins, incumbency, etc show markedly different voting behavior; primaries and the need to motivate one’s base to vote/volunteer/contribute the ideological lumpiness seem to play a large role in deviations from median voter type outcomes).
I’m assuming meta-political discussion isn’t included in the ban on political discussion. Downvote me if I’m wrong, and I’ll abandon this type of comment.
MVT very much applies to my model of U.S. politicians. I see them as having effectively similar policies but very different rhetoric. The similar policies are designed to woo the MV and the differing rhetoric chases the ideological fringes.
That’s a fair description of many parts of U.S. politics (although that sounds more applicable to parties rather than individual politicians: note that foreign aid isn’t a cleanly partisan issue, e.g. Bush massively increased anti-disease aid in Africa, and many Christian Republicans back such aid very strongly), but as I’ll discuss in the relevant post the relevant sort of similarity depends on one’s purpose: the variation among politicians (within, not just across parties) does seem to be enough for a narrowly-focused GWWC member (who can dismiss many standard political issues as trivial in terms of her objective function) to identify differences sufficient for politics to beat the direct effects of donations to the current GiveWell/GWWC charities.
I can think of one big advantage of charity over politics: Giving to politics is rent-seeking behavior. If you give money to support politicians against/for dairy tariffs, then you encourage the other side to give money to their politicians who are for/against dairy tariffs. In this game matrix, “cooperate” is don’t give money.
I have a hard time swallowing the ten trillion dollar figure. I don’t think the two parties are really that different, and there’s good, game theory for them to be that way.
I think there is a market for some sort of organization by which a person who wants to donate $50 to the Democrats can contact a person who wants to donate $50 to the Republicans and mutually agree to donate their collective $100 to some third-party cause like world hunger instead.
I like this idea, but how do you verify that the donation would have gone to the political party? If I’m inclined to give $50 to a non-political-party charity, then there’s incentive for me to claim it would go to the party I prefer in order to get one of my political opponents to divert money from the Bad Guys to another cause.
Agreed.
I would like to come up with aways to prevent gaming this sort of system. If we could get past this sort of hurdle, I would be interested in implementing a website facilitating this.
Perhaps the matched donors have to agree on a charity and an amount: you pay into the system, and if you get matched, then the non-political charity gets your money, and if you don’t (after some waiting period), then the political charity does. This means you have to at least not mind the money going to the political charity, or be willing to gamble.
A thought: Charity popularity probably isn’t evenly distributed between political parties. This could limit the ability of some charities to be matched, but it would allow (for example) a Republican who supports Planned Parenthood to have little or no risk of their donation defaulting to the GOP. A Republican who didn’t want to donate to the Republican Party and did want to donate to Planned Parenthood would take very little risk, while one who wanted to donate to the NRA might have more trouble.
Charity itself isn’t evenly distributed: Republicans give more dollars to charity, period. If everyone in the US signed up for this site, it would guarantee a Republican win. But, of course, not everyone will, so it may not be that bad.
Yes, that should reduce the gaming of the system. It involves more secrecy that in my original conception of the website (which would display graphs of the matching money available for different charities), but I don’t think that can be helped.
Why would my fix necessitate secrecy or prohibit such graphs?
It is probably impossible to become extremely confident about that, but past contributions to a party are a good predictor of future contributions, and I believe that Federal campaigns are required to disclose the names and amounts of any contributions to the Federal Election Commission, which publishes them.
Those published disclosures can be used to qualify a counterparty (a person one is considering making a deal with) and then to verify that the counterparty carried out his end of the deal.
The deal should be that a Republican promises to contribute X dollars less this election cycle than he did last cycle if a Democrat will do the same.
The deal reduces (direct) contributions to campaigns even without the requirement that the Republican and the Democrat agree on a charity to donate 2 * X dollars to. In particular, it reduces (direct) contributions even if the two “sides” of the deal just keep that money.
There are indirect ways to make contributions that do not require disclosure, however, and IIUC these indirect ways are heavily used because they allow individuals to get around the dollar limits on direct contributions. So that consideration brings back the idea that the $X each “side” saves in the deal should go to charity (but I see no need to require that each “side” donate their $X to the same charity, just that the donations can be verified) because it gives some assurance that the $X will not become indirect contributions to a political party, since the total amount an individual is willing to spend on altruism tend to stay relatively constant year over year. But that takes out of the stream of money going to campaigns and parties mostly altruistic money, leaving mostly unaffected the money that expects to profit from the contributions, which might have a bad effect on the political process.
Also, credible arguments have been made that past experience with the dollar-limit laws shows that it is futile to keep money out of politics because the people who make their living from that money have so much influence on the law-making process and because there is so much smart money that wants to contribute to candidates and to parties. Carl’s analysis in the OP of the efficacy of “buying” a vote is evidence for that last point.
Providing there are roughly equal levels of cheating on both sides this sounds like a feature rather than a bug.
The problem is that if the potential participants expect this type of cheating, that their donations will be diverted by someone who would not have donated to their political opponents anyways, they will not want to participate.
Even if that is the case, they’re still getting twice* the leverage they’d get elsewhere.
*Minus the small finder’s fee.
I think you have to just eat this problem. The whole point of the system is that it allows a donation to do double duty as a political force and as a standard charitable donation. It will appeal to anyone who likes both, whichever one is more important to them.
Maybe this should be handled at a different level: the two parties get together and agree to each take a large amount of their donation funds and redirect it to a third-party charity. This seems less game-able.
If the parties could get on board with this, they could provide a separate donation option for something like this. Both parties would then commit to give an amount worth the smaller of the sums gathered into charity. Now for the funders not to donate to the charity option would mean a smaller charity sum for their party and more campaign funds for the opposing party if their party ended up with the smaller charity sum.
It’s not in the interests of the individuals employed by the parties to reduce their budgets, so I can’t see them getting on board with this.
What’s the advantage of either over the other?
For one, it eliminates the problem Alicorn suggests: we already know the money would go to the respective party because it has been donated already. Also, it would be easier to manage, because it would be a single public trade with a large amount of money involved, not many private $100 agreements.
Good points, both. But how would you convince them to do it?
I was thinking that a charity might propose this to each party individually and get them to agree to it. But upon reflection, a better way would be to run a charity (or meta-charity) that works like this: when you donate money, you specify your affiliation, and the charity takes care of donating the difference to the appropriate party.
Perhaps it’s not even necessary to specify a single neutral charity option. You give money to the meta-charity and specify your party affiliation and preferred charity. Then we donate the difference to the party’s campaign fund as usual, and split up the remaining money between charities in proportion to how much was meant to be donated to each.
Vote. Up.
If any other LWers want to start this website, I’ll pull my weight with the marketing.
Wow, I expect this kind of naivete from normal people, but not from LWers. This is exactly the sort of bias-influenced human behavior that LW should be teaching you to understand. It’s more Hansonian than Yudkowskian, but still.
Politics is not about policy. Donors are signaling affiliation. No one will use this service.
Vote. Down.
So you predict also that no one used Nader Trader?
What is your detailed theory of politics that distinguishes them?
Probably true, but that isn’t how donors see themselves. The idea here is to let them achieve the goals they proclaim, while still signaling their affiliation.
It would seem to be aimed at those with multiple signals they wish to send. (Political affiliation, charitable contribution, a general attitude along the lines of efficiency). It seems most likely to appeal to those who would otherwise be directly donating to charities but want a bit extra out of their contribution.
Your conclusion doesn’t follow from the premises (which are well understood by those involved).
What if you give every donor a profile page that shows what causes you’ve defunded? What if you linked your facebook to a chart showing how much you’ve kept away from pro-life/pro-choice lobbyists?
Correct me if I’m wrong, but this seems like an opportunity to signal tons of group affiliation, intelligence, and compassion.
I’d help with documentation/user support.
… yes, I happen to actually like the two most hated programming-related tasks. What of it?
Another potential issue:
If money is diverted from small political donations by lots of individuals, does that increase the influence of big corporate donors over politicians?
Could this effect be ammeliorated by getting politicians themselves to value the cancelling of a donation to their opponents as much as a donation to themselves?
Corporations reap some PR effects from donating to charity, so in theory, they’d be even more motivated than individuals to accept diverted matched donations.
I think public financing of elections is a good idea. If political donations are low it’s much easier for a company to buy political influence through donating money. See Larry Lessig: http://fora.tv/2009/10/08/Lawrence_Lessig_on_Institutional_Corruption
Trying to get people who make small donation to cancel out each other by donating to a third course has the opposite effect of creating public financing of elections.
Detail: the website takes donations in rounds that end on a specified date. Each donation has three parts: a political recipient (say, the DSCC), an offset recipient (say, the NRSC), and a backup charity (say, Village Reach). The backup charity must be scrupulously non-political. At the end of each round, the website finds all the donations which match, gives the political recipient who gets more money the difference in donations, and parcels out each side’s remaining money (which will be equal) proportionately among their named charities.
A $100 DSCC/NRSC/VillageReach donation is thus guaranteed to cause the difference in donations between the DSCC and NRSC to go $100 in the DSCC’s favour; I just don’t know whether it will do so by increasing the money the DSCC gets, or reducing the money the NRSC gets. But either way, if there are donors on both sides then VillageReach will get at least some of the money, and if they’re roughly evenly matched it’ll get most of it.
So the big question is: how could this be made to seem attractive to people? It seems like very few would understand it. “No, give to the DSCC directly - that way they get all the money! Otherwise some Republican will put money in and take away your donation!”
I think this is exactly how to do this. I’m not sure how effective it could actually be, but I’d really like someone to make an honest go for it.
One note: I think there is a motivation for underdog supporters not to contribute, since presumably the less-well funded candidate needs funding more than the better funded candidate. I think this should be relatively small though, at least for races that are at all contentious.
Yes, this problem came up in my discussions.
In theory you could state with each donation how much you’d need to take off the other side in order not to give it to your side and look for matches in a process a little like betting markets, but it’s hard to imagine that working out in practice, so maybe you just aim this tool at evenly matched races.
But politics is not about policy. Political donors want to signal their affiliation with their tribes, not spend their money efficiently to change the world. It’s a modern potlatch. Otherwise they’d be giving to something else in the first place—it’s extraordinarily unlikely that giving to the Democrats or Republicans is the maximal way to impact the world, so anyone who is doing it obviously doesn’t have the goal of efficient charity.
I predict that such an organization would capture < 0.5% of donors by # and < 0.1% by $. That’s a pretty small market—a few hundred thousand.
A brilliant idea.
I wonder which type of voter is more likely to think such a system is a good idea. I would be surprised if both sides were equally open minded regarding such things. (But aren’t American so don’t know which lot is least insane.)
Such a system would appeal to the type of voter who doesn’t vote because he is rationally ignorant and/or calculates it isn’t worth his time (b/c he is not a utilitarian and doesn’t count common benefit as a reason to vote), and wouldn’t donate to a political party because he knows there are far more efficient ways to transform money into changing the expected future of the world. He can see why the system is an efficiency gain using the same tools he can see why voting and donating to political parties is a waste of his time and money.
This creates somewhat of a problem for the proposal, if it is only appreciated by those who can’t benefit from it...
Yes, I think this is a great line of thought (despite Alicorn’s legitimate objection). In light of the fact that there’s a fair amount of overlap in human values even among members of disparate groups, it seems like there’s a substantial arbitrage opportunity attached to redirecting partisan efforts to efforts that are unambiguously good.
Re the Prisoner’s Dilemma element in political competition, for this post I left it out (save the acknowledgement with respect to moral objections, and associated link to the Ezra Klein comments section). More on that to come.
Re the second point, the ten trillion dollar figure is not mine (my estimates and data are coming in subsequent posts). I have an upcoming post on the data regarding the applicability of median voter theorem style analysis in the U.S. (short answer: democratic competition constrains differences between candidates to a small fraction of the variation in outcomes with dictators, but different politicians representing the same electorate thanks to narrow wins, incumbency, etc show markedly different voting behavior; primaries and the need to motivate one’s base to vote/volunteer/contribute the ideological lumpiness seem to play a large role in deviations from median voter type outcomes).
I’m assuming meta-political discussion isn’t included in the ban on political discussion. Downvote me if I’m wrong, and I’ll abandon this type of comment.
MVT very much applies to my model of U.S. politicians. I see them as having effectively similar policies but very different rhetoric. The similar policies are designed to woo the MV and the differing rhetoric chases the ideological fringes.
That’s a fair description of many parts of U.S. politics (although that sounds more applicable to parties rather than individual politicians: note that foreign aid isn’t a cleanly partisan issue, e.g. Bush massively increased anti-disease aid in Africa, and many Christian Republicans back such aid very strongly), but as I’ll discuss in the relevant post the relevant sort of similarity depends on one’s purpose: the variation among politicians (within, not just across parties) does seem to be enough for a narrowly-focused GWWC member (who can dismiss many standard political issues as trivial in terms of her objective function) to identify differences sufficient for politics to beat the direct effects of donations to the current GiveWell/GWWC charities.