Hacker News Reader

Comment by coffeemug | Hacker News Reader

> This seems more of a method for a group of people to profit from a non-profit than it does a new method of actually financing non-profits.

It is both. If the method works, this group of people won't profit from non-profits in exchange for nothing -- they'll get paid for picking non-profits that accomplish their declared goals.

Suppose $X is spent on non-profit donations, and Y% of non-profits don't meet their goals. Then only $X * (1 - Y) is deployed in a meaningful way, and $X * Y is wasted -- it's the hidden cost (Hc) of today's allocation mechanism. So long as the investors get paid less than Hc, contributing $X to the donor fund instead will result in a strictly better world.

I think this is promising. I would definitely consider contributing to this fund.

View on HN

Or shorter: You donate 1$. Of that only 0.9$ go to the cause and 0.10$ is used as premium to insure that goals are met.

I don’t see how the insurers have any leverage to make the non-profits meet there goals. That means that they will probably only pick non-profits with a very good reputation. But then the donors could just give 100% to the non-profit.

Yes, it’s not clear which org is responsible for performance management of each bond, e.g. such an org requires both individual talent/expertise and institutional memory. What incentive does the nonprofit have to meet the target goals, if they get paid in all scenarios?

Here is a 2016 white paper on their process for setting up a fund, https://npxadvisors.com/wp-content/uploads/2016/06/NPX-White...

I'd guess that additional funding is harder if you don't meet goals. A lot of this seems to be pushing towards better measurements and priority alignment in the first place.

> I don’t see how the insurers have any leverage to make the non-profits meet there goals. That means that they will probably only pick non-profits with a very good reputation. But then the donors could just give 100% to the non-profit.

Isn't it now the investors that will make sure non-profits reach their goals? They will theoretically only invest in things that they will expect to meet their goals, so that donors actually pay them back. So donors just have to decide which non-profit's goals they want to support, and then investors help them achieve those goals.

So if I understand it correctly, the primary advantage of this method should be that it helps well-performing non-profits to raise funds. And if a non-profit is high-risk but also high-value (in terms of what donors consider important to the world), they will have to donate more for investors to actually invest - but at least they'll know that they will only have to pay that if the non-profit actually achieves its goals.

Of course, the challenge then is how to properly set the goals and how to objectively measure whether they're achieved.

> Isn't it now the investors that will make sure non-profits reach their goals?

The big question then becomes why should a solely ROI-focused underwriter be better at this type of evaluation than a philanthropic foundation which (i) is actively interested in the goals (and probably also related goals/drawbacks not specified as metrics) to the extent of intending to lose money towards achieving them and (ii) likely to be a source of further funding, which means the nonprofit is more motivated both to engage with them and to actually hit the goals

As an investor, why would I invest in a non-profit with difficult-to-game metrics...

Because juking the stats or a bit of weaselling in the way goals are stated is way easier and more reliable than building things or say, achieving actual health outcomes.

I’m somewhat skeptical.

That is indeed a good question. That said, I don't think the underwriters should be ROI-focused - my interpretation is that current donors would be underwriters. This is a way for them to transfer the risk of what measures will result in what improvements to investors. Evaluation could be as broad as "polio should be eradicated".

But yeah, there's a lot of questions left, and I don't feel like I have a good grasp on what this would entail and all its advantages either.

I hope they are aware of, and have a plan to mitigate, Goodhart's law.

Reply on news.ycombinator.com