It'd also be good if people would write to Wealthfront/Betterment/etc. asking for the ability to donate stock! I e-mailed both companies about it a year or so ago, and both responded that they may add it in the future but had no definite plans. There can be benefits to donating that way, which are explained in this essay by Brian Tomasik. And instructions on how to donate stock to GiveWell are here.

Also, I've been using both Wealthfront and Betterment for a year with similar risk profiles, and Betterment has done slightly better. I know a year's comparison means next-to-nothing for deciding which will be better, but I'm curious to see if others agree with you that we should prefer Wealthfront.

My initial reaction was negative - I wanted GWWC to stay focused on global poverty both due to my personal preference of that cause plus the concern that expanding it could scare off / confuse new people. But after reading the rest of the post I don't have those concerns anymore, and I think this would be a good change. I suspect some of the others who gave negative feedback might have not read the whole thing and have an exaggerated idea of what is actually being proposed.

This is what repledge.com wanted to do, but didn't end up launching due to legal issues. Toby Ord talks about it at around 41:00 of this talk.

I would be really surprised if cost-of-living salary adjustments were near 100% of the increased cost, on average.

The way I think of it is this: Expensive cities typically make companies better able to find and attract talented workers in a close proximity, which is good for business, and they'll have an incentive to pay extra for that. But a city is expensive because lots of people want to live there relative to the space available, and living in an expensive city is somewhat like buying an expensive car. To that extent companies would not pay for the increased cost -- being able to live in that city is part of what you're getting for working at that company.

My only "evidence" that the cost is only partially covered is only from my own job searching in various cities as a software engineer though. I don't know what a wider reading of available data would indicate. But I suspect that cost-of-living is covered less than a simple reading of average salary/cost-of-living data, because part of that correlation will be that the average worker in a high-cost city is better than the average worker elsewhere. Also, if you account for periods of unemployment, being in a higher-cost city would hit harder during those times, which wouldn't show up in salary/cost-of-living data.

One interesting thing is that, in my company, you do get a pay bump for cost-of-living increases if you are hired in, or are moved to, a more expensive location, but if you then move to a cheaper location within the same company, there is not an automatic cost-of-living salary decrease. I wonder if that could be a common practice in many large companies due to sticky wages.

Most investments are probably pretty replaceable though.

By this, do you just mean that the selection of one investment over another usually doesn't matter? Or are you saying overall investment is replaceable in some way, i.e. my increased investment can lead to someone else investing less?

As part of the pros/cons between "give now" and "invest now, give later", has there been any investigation into how much good is accomplished by investing itself? It seems like that is a (small) part of economic growth and innovation, so I'm curious if there's much reason to think that has a big enough impact to include in the invest-then-give decision.