This is part of a series of blog posts examining seven arguments I laid out for limiting Good Ventures funding to the GiveWell top charities. My prior post considered the third through fifth arguments, on influence, access, and independence. In this post, I consider the sixth and seventh arguments:
Argument 6: If no one else is willing to fund a program, then this is evidence that the program should not be funded. Crowding out other donors destroys this source of independent validation.
Argument 7: If Good Ventures fully funds every high-value giving opportunity it finds, this could lead to other donors preemptively abandoning programs the Open Philanthropy Project is looking into, thus substantially reducing the amount of effective giving in the Open Philanthropy Project's perceived current and potential focus areas.
Argument 6 is sometimes an important consideration, but is a poor fit for the GiveWell top charities, to the extent that most donors are already largely moved by GiveWell's recommendations. Argument 7 points to a real problem, and one that reflects poorly on the effective altruism movement, but the problem is not mainly concentrated in the area of funding. Continue reading
The 2016 US presidential election is likely unusually important, because Trump seems unusually likely to damage global coordination in ways that increase the risk of major wars – and to damage US political norms in ways that are likely to accelerate the decline of discourse and governance.
This is also an election in which the libertarian candidate has been unusually viable because he has any experience at all as a major government figure – despite his apparent lack of interest in the sorts of things a president needs to know about, such as other countries. Many people also want to register a protest vote with Green Party candidate Jill Stein, since they find Hillary Clinton's respectable establishment liberal misconstruals of the true and the good objectionable, and prefer disrespectable anti-establishment left-wing misconstruals of the true and the good.
Many people agree that Trump is terrible and it would be much less bad if Clinton wins, but some people prefer a third-party candidate and are unwilling to simply back the lesser of two evils. Some people end up favoring a vote for Clinton on net; others favor a third party vote. Both types are distributed over many states.
A protest vote has the same value anywhere. Federal funding also becomes available for any party that gets more than 5% of the popular vote – and it seems like Johnson's share of the vote could pass that threshold. On the other hand, due to the US electoral college system, the cost of forgoing a Clinton vote has very different effect depending on which state you're voting in. In a "safe state" overwhelmingly likely to go to one of the two major candidates, your vote has very little effect on the outcome of the election. But in a "swing state" where the outcome is more in doubt, your vote has a comparatively large effect on the outcome. Scott Aaronson points out that this distinction creates the opportunity for gains from trade, and has been promoting the idea of vote-swapping in order to reconcile these interests. The idea is that one or more Clinton supporters in safe states pledge to vote for a particular third-party candidate, in exchange for a third-party voter pledging to vote for Clinton.
In a one-to-one swap, this keeps third party national percentages the same, but increases the chance the swing state goes for the desired candidate. This is enough to yield gains from trade if both sides share a preference for one major-party candidate over the other. But even if that's not true, a many-to-one swap can still create gains from trade, by increasing both the chance that the desired major-party candidate wins, the third-party candidate's share of the vote total.
One of my friends recently suggested that we can't trust this system not to be gamed by Trump voters. I think that this is mistaken. Continue reading
It took me a while to figure out how to write regularly. I had to do a lot of iterated troubleshooting before I figured out how to reliably generate output. It's possible that none of those are the insight you need, or that internalizing them is mostly not about having the words for the thing - but I figured it was low-cost to share anyway. Plus I want to write this up and now seems as good a time as any to put it in writing. Continue reading