Friday, November 27, 2009

Economics of small donors--Reverse Auction Charity Shares

In spite of the collapsed connectivity of all things, it's a little hard to explain why I'm filing this one under the general category of epistemology... In a sense, economics is a kind of reality underlying the real world, and the epistemic purpose of philosophy is to understand real world to improve it, but this is also a kind of 'business proposal' to make the world a better place. Another reason is that the various aspects of this proposal stick out into various other fields of endeavor in weird ways. Ergo:

The working title of this new economic model is "reverse auction charity shares". Of course it can be mutated and improved in many ways, but I think this label captures the essential features. One way to think of it is as a different kind of stock market that allows for small donors to have some deeper visibility into what they are donating for and how their donations are used. A large donor is in a position to shop around for charities and can negotiate the size and conditions of each donation to satisfy various goals, but right now the smaller donors are much more limited in their options.

Interestingly, there has been some evolution in the direction of this proposal, since many organizations that are soliciting for charities these days do try to target the donations for specific campaigns or objectives, but there is still very little visibility there. The current situation for small donors is more like a form of email marketing. As a result of signing up for mailing lists or based on prior donations, you (if you're like me) receive a steady flow of solicitations for one thing or the other--but the organizations persist in acting like it's a sellers' market, while the goal of this proposal is to change it more to a buyers' market of the same sort the large donors already have.

First, the key term "projects" needs to be defined. For this proposal, projects should be taken quite broadly. This basic mechanism can be used for many purposes such as developing open source software, publishing Web magazines, supporting political campaigns, a class action lawsuit, or various other goals. The basic idea of a project is some defined unit of work that can be accomplished if enough people want to donate to pay for it.

The "charity shares" part is relatively easy to explain. There would be 'virtual shares' in 'donation ownership' of each project. Though the shares would not confer real ownership, the donations would be publicly recognized (if the donor wished) on the webpage recording the donations to that project. This would actually be a view into a donor database for those donors who wanted to be publicly recognized for their association with a particular project. A large project might have thousands of donations, but I think there will be more interest for relatively small projects that are easy to understand. Those smaller projects will also give better visibility to their donors, giving them more significant recognition and a larger feeling of accomplishment for their donation.

From the perspective of a charity stock market (presumably implemented as a website), the donors would be able to shop around for the kinds of charities they want to support, and then buy shares in those projects. The projects would not actually commit and begin work until enough donors had purchased shares, but the funding status of the project can be visible to all of the donors at all stages of the process. The charity stock market can also be visualized as a kind of database search, where the prospective donors are indicating their goals and the website is recommending the kinds of projects that are available to meet those goals, including such factors as how close those projects are to committing.

There are other possible inducements for these virtual shareholders. For example, the donors can help evaluation the results of the projects that they support, considering several dimensions of success. Another obvious wrinkle is to offer shareholders opportunities to invest in similar projects as they appear. From a marketing perspective, there are possibilities such as matching shares from corporate sponsors and donor pairing or various other approaches.

The "reverse auction" part is a bit harder to explain. It is based on the timing constraints, which are always important, while adding a kind of 'viral' incentive for donors to recruit more donors. Essentially if the donors can recruit more donors, they would still accomplish the goal that they already felt strongly enough to donate to--but at the same time they would drive down their own cost for accomplishing that goal. The essential idea here is to align the incentives to maximize participation.

The approach would be to sell excess shares beyond the initial allocation while cutting the share prices to reflect the additional support. A proposed project budget might call for 1,000 shares at $50 each. There would also be a target date regarded as the closing of the auction. However, by selling extra shares, the per share price would go down, even though the project was already committed for implementation. The donors would be motivated to seek more donors from their like-minded friends, perhaps selling another 1,000 shares while driving the per share price down to $25. If the final limit on shares is 2,000 shares, then selling out completely would be a strong reason to expedite the implementation (and to consider creating additional similar projects).

There are various possible mechanisms to handle the declining share prices for the earlier donors, but I think the simplest one is to use a kind of 'charity brokerage account' system. The donors would have a balance with the brokerage, and they would commit their money to various shares for the projects they liked, but the money would not be deducted from their balance until the final closing of the deal for that project. This would also serve to create an ongoing relationship with the donor, and create possibilities for such things as conditional share sales. For example, a donor might commit all of a $100 balance to buy shares in 4 projects, and buy a conditional share in a 5th project. If the closing prices of the first 4 projects decline enough, then the donor would be able to join the 5th project, too. Again, the donor has a positive incentive to encourage other people support the same projects.

I hope an example at this point will clarify more than it will confuse, though each focused example will introduce new wrinkles and possibilities. Consider the case of a magazine with a Web-based version. Acting as the charity brokerage, the magazine would be equating donors with subscribers, but these subscribers would be buying the right to buy shares in supporting various authors and articles published by the magazine. In this situation, to exercise editorial control, the publishers would probably adjust the stock prices to reflect their editorial priorities. For example, they can deliberately create high prices for supporting a popular columnist while simultaneously offering less expensive shares in related projects that can attract some of the subscriber/donors in this application. For a popular article that sells out both the basic and extra shares in the Web version, the publisher can offer a special issue of shares to support publishing the same article in the print edition, essentially substituting some Web-derived revenue for conventional advertising revenue. In that case, donors might be motivated by wanting to support the wider publication of authors or positions that they agree with.

However, the most important aspect is really that the economic model allows for many kinds of experiments in marketing ideas to the real audience, while simultaneously letting the audience vote with their wallets. This is not intended to eliminate other funding approaches, but I'm suggesting it as an additional donation mechanism, with various kinds of side effects. For example, one interesting side effect is that the visibility of this mechanism would be highly resistant to astroturfing (as when a project has too many anonymous donations). Corporate donations can be incorporated in several ways, such as matching share purchases or up-front donations that reduce the share costs to the other donors. In addition, donors would be able to look over their own donation histories to assess what kinds of projects they had supported and get recommendations for similar projects they might be interested in buying into.

Let me know if you're interested, want details about other possible application areas, or have any suggestions about where to pitch this idea. I've already been flogging it for a while now, and as far as I know, it has not yet been adopted.


  1. Not much sign of interest so far, eh?

    Anyway, in applying this kind of model to software development (especially thinking about Ubuntu's deteriorating situation), I think that the project models should have substantial allocations for testing in their budgets--but the virtual shareholders would also become the highest-priority candidates to become testers. In other words, if you bought a share in a particular project, you could also volunteer as a tester at that time, and then record your configuration as a test candidate. (Or you could better your odds if you have several machines.) At that point it would be a bit of a lottery, but essentially the project managers would be picking testers to maximize test coverage (and with some actual knowledge of the kinds of machines the donors have), and the winners would be paid for their testing work. The value paid for testing would probably be more than the cost of shares, so the winners would be extra happy and it would be yet another motivation to support a development project.

    I think the main advantage is that the projects would be buying testing from normal users--but perhaps that's because I think that the earliest users of most programs tend to be on the strange side or stranger. The main risk might actually be a kind of gambling by people who think they have especially useful configurations for testing...

  2. Another wrinkle recently occurred to me... A project could commit and start work as soon as sufficient shares had been sold, but the extra shares and per-share discounting could easily continue after that, perhaps until the actual completion date of the work. Since the money is already in the charity brokerage, it is effectively in an escrow status, and the final per share price can be determined later.

    This might actually be a marketing consideration. It might be a stronger incentive to buy shares in a new project if you receive a combined notice that your current project is completed AND it 'cost' you even less than you thought. (Using 'cost' with the quotes because you've already donated your money to the brokerage, and you had already decided the project was worth supporting at a higher per-share cost.)


    This is a related idea that was recently called to my attention. However, I think that the system has several flaws compared to my version. I actually spent some time on their website today trying to assess the results, and I couldn't find much. Perhaps I was looking at the wrong parts of the website?

  4. Hi! I like your idea. I'd like to see a formalism of it, but I like where you're going with it. I have no additions/full criticisms to offer yet, but I thought I'd voice out some praise. Keep up the creative flow!

  5. I hope the moderation wasn't too slow, but I really hate the spammers, so I have to stand by it...

    As far as "a formalism" goes, I'm not quite sure what you're asking for. A spreadsheet model? A sample project budget?

  6. Something of this model seems interesting but it's too complicated. Try to simplify it so that it can be explained in a few sentences instead of that much text.

  7. Well, the short explanation is that it's basically a mapping from a stock market with a broker to a system to support charity. The donors don't get dividends, but they are listed as 'virtual owners' who supported the project. Most of this idea is almost in a one-to-one relationship with a stock market except that it is fundamentally a charity, so the donors don't take money or profits out. The money goes to the expenses of the charitable projects, but only after a sufficient number of donors agree the project is good.

    The main variation that makes things a bit more complicated is the idea of overselling the shares. This is a kind of viral marketing aspect so that it is always in a donor's interest to encourage other donor's to join in supporting the project. Initially, before reaching the initial number of shares, the donor is striving for the commitment to start the project. After that, in the overselling part, the donor is reducing the donor's own cost of supporting the project.

    If you have a more concrete question, I can try to answer it.

  8. Another idea came to me about contingency funding. The project proposals should include consideration of possible problems and the likelihoods of various unanticipated costs, and every project should include a contingency fund. Perhaps 10% would be normal?

    But what to do if the project is well executed and they don't need the contingency fund? Then that money would revert to the brokerage and become 'bonus' funding for other projects. One obvious way to allocate the unused money would be to let the first shareholders (from the shares in the completed project) each get a chance to buy a share in some other project, going down the list of shareholders until the money is committed.

    Obviously, this can be pitched based on projects related to the project they just supported, if they can consider their other interests it provides another kind of useful feedback. From that perspective it also becomes another aspect of the marketing for projects. At the same time, it also encourages people to be early shareholders and to focus on projects that seem to be well managed and more likely to succeed without expending their contingency funds.

    This idea will also encourage planning ahead. If the project's creators want to do a follow up project, they should have that proposal ready when they finish the current project. By doing this, they have a good chance to get their most enthusiastic supporters (the early shareholders) to commit the contingency funds to the next stage project.

    In general, I think the lose ends should tie together if the overall concept is sound. This is just another example of where there's a natural fit to the funding needs.

  9. Mostly thinking of these comments as notes for the next version of this idea. All of the components need to be reconsolidated, eh? Anyway, several new ideas today, though I'm not sure of the sources... The recent passing of Steve Jobs? The book called "The Hacker Ethic"?

    Anyway, my original idea here was that the programmers would bid below their market rate in exchange for the added control that allowed them to work on their interests. However, now I see several rationales for bidding at higher rates. One would be an appeal to fans of the programmer's style in previous projects. The other would be a reuse-based rationale. If a programmer writes in such a way that many other programmers like to reuse his code, then that's an argument for higher rates as an important innovator who supports other programmers with leading-edge code.

    That last idea leads to the notion of including a code-reuse budget in the project budget. I've started thinking of what components should be in the basic template of the project budget. The testing budget is probably the most important to me, but kicking back some money from a successful derivative project should encourage upgrade and enhancement projects. In other words, programmers could provide customized versions at relatively small costs, where the project would involve only a few hours of programming, some sort of royalty payments to the main code sources, and (of course) the testing budget.

    On the notion of eliciting code contributions, there should be a budget to encourage that, too. I realized this could almost work like a lottery. If the external-programmer budget is allocated based on relative contribution, it would be a good strategy to study the code enough to make some minor improvement somewhere in the code. If there were no other contributions of code, then that entire budget would go to the person who made the minor improvement... Not sure if that's a good thing or not.

    I've also started developing some related ideas on the Aristotelian Principle, but this really belongs with a review of the book...

  10. One more idea for enhancing this idea in the context of publishing. Before, I suggested exercising editorial guidance by manipulating to share prices, but here's another way editors can steer things.

    The basic idea what is to award little bonus stars that would appear next to the names of donors who had bought charity shares in favored editorial projects. For example, a precious gold star could be for buying into a promising project before it is written. A silver star could represent buying a sponsorship charity share after the article was written and looked good, while a bronze star would be for latecomers who just wanted to buy special sponsorship after it won an award, such as a Pulitzer Prize.

    In addition to expressing the editorial objectives of the editors to do good journalism, these funds would be more discretionary and could be used for more speculative projects that the readers didn't pay initially want to pay for. In other words, the editors could fund some projects that they favor even if the readers disagree--as long as they produce some results as defined by whatever criteria qualify for silver and bronze stars, whereas the gold stars would represent stronger agreements between the ideals of the the editors and the interests of the readers.

    Having said that, I still don't see any sign of this idea moving into the real world...