26 Comments

Regarding 12a: How should a young scientist go about marketing shares of future prizes? I am a graduate student and I feel that basic science research is painfully underfunded. This is not to say that "science in academia" is underfunded; on the contrary, there is a lot of money available if you want to do research that caters to short term commercial gains for companies and sponsors. As a scientist, I want to specifically avoid this kind of research from the very start of my career (most advisers tell me to just "play the game" and try to sneak in a couple of 'basic' research papers here and there when I can eek out the extra time. To me, this is a poor solution and if I can't do better then I'll just leave academia and embrace the fact that I have to cater to short term commercial interests (which are inherently very uninteresting to me) and at least earn some money).

You seem to have a successful track of publishing research that's tightly clustered around your interests. How can a young (not yet established) scientist do the same thing? I don't consider having to "conform to the system" until I become "established" to be a feasible answer.

Expand full comment

And to pre-empt any mention of long shots that end up being extraordinarily beneficial: don't worry about the people working at the research lab that cures cancer. I have a feeling they'll be able to put food on the table.

Expand full comment

So your charity model is to pay people for work?

So like, let's say I can buy a burrito for $5 and eat it. While I'm at it, I can give $5 more to give the broke hungry man next to me a burrito.

One could claim that I've bought a $5 burrito and also given a $5 cash prize to the burrito store for an act of charity. One could also claim I bought two burritos for $10.

This sounds like exactly the kind of plan an economist would come up with.

But how about we go one step further and focus specific organizations to altruism? After all, it's usually harder than buying a burrito for a man standing next to you. The beneficiaries and donators are probably going to be quite distant; that's why you need the organization. To try and ensure a minimum of fraud and corruption, these organizations will have to be financially transparent and avoid profit-seeking. We'll call these "non-profits". So the idea is, you'd give these "non-profits" some money, and then they would do a specific service that is a matter of public record. People would be able to donate to non-profits that they deem to be effective investments of cash.

So, if I may finally shed the veneer of sarcasm, you basically implied a method for seeking out effective charities ($5. 1 burrito. Clear cost and clear effect.), without actually suggesting any kind of new institution.

Expand full comment

As I discuss below responding to Holden, underfunded doers is an exceptional case. If they must promote their potential, that just brings us back to a grant like level of promotions for that exceptional subset of cases.

Expand full comment

Jess, those young intellectuals aren't going to get much grant money in today's system. So hard to see how prizes make it worse.

Expand full comment

Interesting! Reading Holden's and Hanson's points, I thought of GiveWell as a proxy prize. If GiveWell was more well-known, such that a good recommendation from GiveWell causes an increase in donations (is this the case even now?), that increase can be considered a prize for passing GiveWell's standards.

To the extent that GiveWell's standards are a good approximation of "achieved a lot" (part of their evaluation is looking for statistically significant improvements from the charity's work, which is precisely "achieved a lot"), they are a specific instantiation of Prof Hanson's recommendation - that sacrifices a small amount of efficiency relative to the recommendation to gain a large satisfies-other-goals-of-charitable-giving bonus.

Given that it appears many people have other goals for charity (see "But if you think yourself one of the rare exceptions"), I think the tradeoff is a good one.

Expand full comment

This seems incompatible with letting underfunded dogooders get investment. "Fund me and it will be so awesome people will donate." seems extremely speculative unless this kind of donation were common. This could be facilitated with a website devoted to such donations, like a time-reversed kickstarter.com

I also wonder if people will be less willing to donate if they know much of the money will go to investors instead of "the people who really did the work"

Expand full comment

most intellectuals do have some discretionary resources to allocate, especially most of the ones you would actually want to attract to work on your cause

My first instinct is that many important causes would benefit most from drawing young intellectuals (i.e. grad students and post-docs) who currently squeeze into over-populated fields because, though crowded, these fields guarantee an audience and source of citations for mechanically produced papers. New and underdeveloped fields are too high risk for many young academics unless they are guaranteed steady employment for at least a few years and a chance at tenure.

Expand full comment

I'd rather people spent time promoting what they've actually done than spent time promoting what they think they might be able to do if given sufficient grants.

Expand full comment

Holden, first most intellectuals do have some discretionary resources to allocate, especially most of the ones you would actually want to attract to work on your cause. Second, yes in general it might help for customers to band together to promise large purchases if certain products were developed, to help entice investors to try to develop those products to sell to such customers later. But there is certainly no general requirement for such customer group commitments in order for capital markets to induce and fund investments in pursuing such customers. A great many products are developed without such commitments. This general point applies to future charity angels as customers of intellectual's ideas, as well as to many other kinds of products and customers.

Expand full comment

There are plenty of prize sites that would do a lot of the advertising for you. If you put up a competition on kaggle with a few hundred dollar prize you can get many people working on a prediction model. The first challenge on their page at the moment has a $150 prize.http://www.kaggle.com/

I am not sure what predictions models would be useful for charity? Maybe a model that would predict the number and location of cases for some infectious disease?

There are other challenge sites out there with their own communities so i think the advertising problem is not as bad as you might imagine.

Expand full comment

Robin, I don't see how you are addressing the argument that grants are important for the capital they provide, not the incentives. I see a grantmaker as trying to find people who don't need any more incentive to accomplish good things, but do need more capital. Grants provide capital; the "charity angels" idea doesn't.

Certain incentives can be "converted" into capital. For example, people know that certain kinds of businesses reap rewards; therefore, there are investors who provide capital to businesspeople, in the hopes of later getting a cut of these rewards. But this is where my point #4 comes in - incentives have to be psychologically prominent & seemingly predictable enough to build a plan around and be worth taking a risk on. I think the X Prize succeeds at this, and that a widespread enough "charity angels" program might as well, but that individuals acting on your suggestion are unlikely to create prominent and (seemingly) predictable incentives for doing good. I wasn't referring to economies of scale in charity, but rather in prizes.

Expand full comment

If you want to say things other than money also motivate us, then I'm all with you.

But I don't think you want to go to the other extreme either. The video says money doesn't motivate people for cognitive tasks at all. That's even more silly.

Expand full comment

Holden, on 1, that is a big advantage of prizes over grants, especially if you wait longer before awarding them. On 2, that is a reason to give to individuals instead of organizations. On 3, monetary grants and prizes are equally ineffective at motivating folks who don't want money. Once you have something to offer that they do care about (e.g., sex), then the issue of rewarding promise before hand or accomplishment afterward becomes relevant. On 4, you suggest there are scale economies in charity, but don't really say what they are. Yes of course if there are scale economies whereby big amounts more than proportionally effective, then donors should try to coordinate to join together to offer bigger amounts. But coordination is expensive, so I'd like to see a clearer case made for the existence of such scale economies.

Expand full comment

This is Holden Karnofsky, co-founder of GiveWell. We try to help people do as much good as possible with their giving, and we're open to anything - if we became convinced of the merits of this idea, we would promote it. However, we are currently not convinced. A couple of points:

1. It often takes a long time to understand the impact of the kinds of initiatives we're talking about (i.e., initiatives that set out to accomplish good in a way that is not profitable / does not pay for itself).

2. If you give to an organization that has accomplished great things, you may be "rewarding" completely different people from the ones who were responsible for the accomplishment.

3. I believe that people who set out to accomplish good in a way that is not profitable are usually confident in / happy with their financial situation, and looking for a non-monetary reward. Donations to people who have succeeded may thus have diminished impact on incentives relative to most monetary transfers.

4. The points above are assuming that "charity angel" activity is to be undertaken in a fairly small-scale, irregular way. Visible, regular rewards for good works may work well, especially if they were predictable enough to allow people to borrow the needed resources against the hope of these rewards. Therefore, if you saw an opportunity to promote this idea in a big way and turn it into something visible and regular, I'd be more supportive of that; I think it is a poor idea for scattered individuals trying to accomplish as much good as possible with 3-6 figure donations.

Expand full comment

It's much more efficient to pay the producers of ideas directly. Universities, and even the most efficient research groups, produce WAY fewer ideas per dollar than, for instance, Robin Hanson does.

Expand full comment