I'd like to start an organization that, similar to the website Patreon, enables people to set financial rewards on mathematical proofs of their choice and holds their money in escrow.

Since some proofs take decades to be written (and some might be impossible to write), this would have to be a pretty long-lived organization, protected against the failings of individual humans (theft, death).

If this were about a single person awarding price money for a single proof, a will might do. But this organization should enable many people to contribute money and objectives.

Ignoring issues like inflation and the complex and possibly costly process of proof verification for now, what kind of organizational form would be best suited for this undertaking?

A Charitable trust?

A Nonprofit organization?

  • @Justaguy Re. country: US or Germany, or others if more suitable
    – 2080
    Aug 13 '20 at 15:36
  • I assume you've seen the Erdos solution: Instead of getting a foundation, get a good friend who is younger. quantamagazine.org/…
    – Just a guy
    Aug 13 '20 at 18:59

You can do it using a US non-profit, but you need a lawyer

We know it can be done in the US using a non-profit because it has been done. For example, the Clay Institute’s Millennium Prize does exactly what you propose: it pays people who solve unsolved math problems. Similarly, the the Everglades Foundation used a prize to get people to come up with new ways to get phosphates out of the water sources.

To do what they did, you need to follow their lead, and set up and run a non-profit (aka, 501(c)(3), named after the relevant provision of the tax code).

Because non-profits are tax exempt under US tax law, you have to apply to the IRS to become a non-profit. Speaking broadly, the IRS imposes two requirements on non-profits: 1) It must serve a charitable purpose; 2) It must spend a certain percentage of its money every year doing charitable work.

To set up a foundation you have to specify the charitable purpose you will serve (“encourage research in mathematics”), and have a plan to do so (“we will award a cash prize to people who solve important unsolved problems”). If the IRS decides your plan is genuine, then they give you a pass on paying taxes.

To make sure you are carrying out your charitable plan, the IRS requires you to spend a certain percentage of your money each year on charitable work. (This is called the "distribution requirement.") Meeting the distribution requirement may be a problem, since you may not award prizes every year.

You may be able to get a general idea about how the IRS deals with such situations by searching on line. At some point, however, you will probably need to talk to a lawyer with experience setting up non-profits. (FWIW, it looks to me as if the Clay Institute meets its yearly spending requirement by supporting lots of other activities that count towards the spending requirement.)

One other problem that it seems you won’t have involves giving prizes directly to individuals. For obvious reasons, the IRS is generally suspicious of prizes given to individuals, so they impose restrictions. For example, for the winner to avoid taxes, the prizes have to be for previous work.

Where to go for more information:

USA.gov has a nice overview of the application process here. The IRS has several publications and websites that you might find helpful: here, here, here, and here. These cover everything from the general rules to nuts-and-bolts details about which forms to use.

You can read more about the IRS treatment of prizes here and here.

Depending on how comfortable you are with financial statements, you may learn about how these non-profits work by reading their 990 forms. The Clay Institute's forms are here.

  • 1
    Thank you very much for your answer! One cannot seem to avoid significant paperwork when trying to establish something that will have its own legal existence :) I suppose this poses a bootstrapping problem, where setting up and maintaining such an entity requires significant resources, which only exist when the network effect for acquiring them has already been achieved, which is, I guess, why this doesn't exist yet.
    – 2080
    Aug 13 '20 at 18:25
  • 1
    @2080 Sadly, I am afraid you are right. Either you pay a lawyer who has already spent a lot of time learning the law, or you pay by spending your own time to learn the law. But either way, you pay. PS You are welcome.
    – Just a guy
    Aug 13 '20 at 18:57
  • 1
    @2080 Some other options: 1) Find a rich(er) donor to fund the startup costs (you could offer to name the first prize after him); 2) Piggyback on an existing foundation; or, 3) Find a smaller foundation that prizes (ie, Everglades) and figure how they solve the distribution problem. (If you are lucky, they might even be willing to talk to you.)
    – Just a guy
    Aug 13 '20 at 19:03
  • 1
    @2080 If I under ohwilleke's answer to your other question, and if he is right, then distribution is not the problem I thought it was. According to him, the bulk of the money remains in the estate, and the foundation (or special master) simply decides when the terms of the will have been met so the money can be dispersed from the estate. (Of course, as practical matter, the only difference may be that he would have you pay a trusts and estates attorney instead of a tax attorney.)
    – Just a guy
    Aug 13 '20 at 19:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.