What is Crowdfunding?

Crowdfunding is a fundraising strategy that encourages individual members of some “crowd” or group, such as the public, to contribute money, often variable amounts of money, via the Internet, to produce a specific good or service. According to Lambert and Schwienbacher, crowdfunding is an activity involving “an open call, essentially through the Internet, for the provision of financial resources either in the form of a donation or in exchange for some form of reward and/or voting rights in order to support initiatives for specific purposes”. Lambert and Schwienbacher define crowdfunding as a kind of crowdsourcing. Like crowdsourcing, crowdfunding allows the “general public (or the crowd)”, as individual “consumers”, to voluntarily “provide input to the development of the product, in this case in the form of financial help”.

Crowdfunding websites allow their members to seek funding for their own initiatives, as well as fund the initiatives of other members. In this way, crowdfunding invites users to act as both fundraisers and funders. To seek funding from the crowd, members specify a fundraising goal, including information on who is seeking funding, how much money they need, what they need the money for, when they need the money, and what rewards (if any) funders will receive in exchange for their financial support. Rewards include rights to use or own what is produced by the initiative, such as the right to own a copy of a song produced by the initiative, or rights to use or own something that is an accessory production, such as ownership rights to a thank-you gift.

After a fundraising goal has been published, other members of the website can then review the current status of these fundraising initiatives. The current status of a fundraising initiative often includes information about who has currently given funding to the initiative, how much they gave, how much funding is still needed, and how much time is left to give it. By associating a funder with what they fund, funders are rewarded with social credit for what the initiative intends to produce with those funds. In this way, most crowdfunding websites minimally serve as markets that facilitate the exchange of money for reputations, and maximally serve as markets that exchange money for both reputational and material rewards.

There are different kinds of crowdfunding websites that offer different kinds of financing, including grants, equity, and loans, each with different kinds of rewards. Websites like Kickstarter allow users to crowdfund project grants in exchange for rewards determined and offered by the fundraiser, while websites like Kiva, allow users to crowdfund microloans in exchange for social kudos and eventually an interest-free repayment of the loan, which can then be recycled into other microloans. For our purposes, we will mainly consider crowdfunding grants, since grants are a common form of financing research.

Examples of Crowdfunding Research Grants

Kickstarter is a crowdfunding website that provides a platform “to fund creative ideas and ambitious endeavors”. It uses an “all-or-nothing” mechanism that allows funders to pledge money toward a project, but it only transfers money from the funders’ pledges if the project’s fundraising goal is met by a certain deadline. This is supposed to lower the risk of project failure because all projects that receive funding are fully funded. Since a project is less likely to fail due to inadequate funding, people, including funders, are more likely to realize the benefits of the project. Kickstarter is a platform for crowdfunding grants because funders retain ownership rights over what they produce, and only optionally provide rewards to funders for contributing specified levels of funding. Kickstarter funds a variety of creative projects, including research projects. However, it has rarely funded research projects.

By searching for the keywords “research”, “grant”, and “study”, it appears that Kickstarter has only successfully crowdfunded a few research grants, and most of this research involves multimedia deliverables. For example, Mary Maxwell successfully used Kickstarter to crowdfund her research on plastic pollution in the ocean. Fifty-five people contributed, in total $5,000, to finance Maxwell’s “voyage from Bermuda to the Azores, as well as the cost of plastic sediment analysis, fish tissue sampling, and marine debris density research.” To motivate donations, Mary relied on a mixture of material rewards and visual rhetoric. To engender sympathy from potential funders, she showed a picture of a deformed turtle whose shell had grown around a piece of plastic trash. To entice funders with material rewards, Maxwell offered “a handwritten postcard from the trip” to those who offered $25 or more; “a personal thank you letter and an autographed photograph taken on the trip” to those who offered $50 or more; “an autographed book of photos, journal entries, statistics and findings from the trip” to those who offered $200 or more; “a teaching kit consisting of photos, journal entries, statistics, findings from the trip, slide show or video footage, water or sediment sample, and suggestions of actions you can take to help solve the problem” to those who offered $500 or more; and “a weekend stay for two in luxurious hotel accommodations in San Francisco and a live presentation about the voyage, in addition to a teaching kit consisting of photos, journal entries, statistics, findings from the trip, slide show or video footage, water or sediment sample, and suggestions of actions you can take to help solve the problem” to those that offered $1,000 or more. In fact, 1 person donated $1000 or more, 2 people donated between $500 and $1000, 3 people donated between $200 and $500, and 18 people donated between $50 and $200, and 14 people donated between $25 and $50. One research question that we might ask is to what extent does the success of crowdfunding research grants depend on offering multimedia rewards like pictures, audio, and videos of that research. If the success of crowdfunding research grants significantly depends on offering multimedia rewards, then crowdfunding may not be a viable way finance research that does not lend itself to visualizations.

One relatively unique design of Kickstarter is how it handles social credit. Unlike many other crowdfunding platforms, Kickstarter does not publicly indicate the amount contributed by each funder, and so does not give more social credit to to those who give more money. Kickstarter only lists the usernames of the funders, links to funder profiles, and the dates on which they were funded. Presumably this is to protect the economic privacy of the donors; perhaps certain donors don’t want to disclose the magnitude of their wealth and generosity. This opens another research question for crowdfunding: Does the display of contribution amounts significantly influence the likelihood that a funder will decide to contribute money?

PledgeBank is another crowdfunding platform that has been used to fund research grants. Unlike Kickstarter, it does not collect money, or limit itself to financial pledges. Instead PledgeBank allows people to post conditional pledges “of the form ‘I will do something, if a certain number of people will help me do it.’”. For example, in order to crowdfund Myalgic encephalomyelitis research, Serena Blanchflower posted the following pledge to PledgeBank: “I will pledge £250 to MERGE, to fund much needed biomedical research into ME but only if 25 other people will give £10 each.” PledgeBank allows other members to sign the pledge, which thereby pledges them to meet the condition of the original pledge. If the requisite number of people sign a pledge, and agree to meet its condition, then PledgeBank notifies everyone that the pledge has passed, and they should follow through with their pledges. PledgeBank allows people to indicate whether or not they have acted on their pledges, but it does not attempt to enforce it. In particular, it has no mechanism for collecting or transfering pledged money. The PledgeBank website claims that they have “surveyed many of the money-based pledges that have succeeded, and found that payment rates vary from 50% to well over 150%, with three-quarters of people paying being typical.” Unlike Kickstarter, PledgeBank does not emphasize a total fundraising goal amount, but instead focuses on the total amount of participation. For example, Blanchflower’s pledge does not emphasize that she is looking for at £250 of donations and if that she would match it with £250. Instead, it emphasizes that she is looking for 25 people to participate by donating £10 each. This raises another practical research question for crowdfunding: Does emphasizing a total fundraising goal amount tend to lead to significantly more total donations than emphasizing total participation?

In thinking about how to crowdfund academic research, it may be helpful to think about how investigative journalism research has been crowdfunded. Crowdfunding platforms like Spot.Us demonstrate some success in crowdfunding research grants for investigative journalism. For example, Peter Byrne used Spot.Us to crowdfund his investigation into how some members of the Regents of the University of California have personally profited from how billions of dollars of public university monies are invested in private companies that they own.

Spot.Us allows users to both pitch stories as citizen journalists and fund pitched stories as media consumers. So far, Byrne has raised $5,037.00 of the requested $10,000 from 83 people. Byrne’s fundraising deadline is 8 days overdue, but unlike Kickstarter, Spot.Us does not use an all-or-nothing model of crowdfunding. Spot.Us allows fundraising journalists to decide to accept a lower amount of funding than what they had originally requested. However, Spot.Us only pays journalists after they have produced their stories.

Spot.Us is also different from Kickstarter in that it does not offer individual rewards to funders. It only offers individual rewards to media organizations that are willing to finance at least 50% of a news article. In that case, Spot.Us temporarily gives those corporations exclusive copyrights to the crowdfunded article. In order to receive funding, Spot.Us requires freelance journalists to sign an independent contractor agreement, which gives Spot.Us “the sole discretion to decide whether, when and how to publish the Article and may sell or license any rights to the Article.” The agreement also notes that “it is the intent of the Organization to eventually release the content of the Article under creative commons.” In this way, funders of an article expect to eventually have the right to read an article, and let others read it for free. The independent contractor agreement also grants Spot.Us the right of first publication, preventing the freelance journalist, within the first 90 days, from publishing the content without Spot.Us’s permission. Spot.Us also requires would-be journalists to work with a peer-reviewer that Spot.Us chooses (although it allows the reporter to reject the first two editors that Spot.Us selects). One major problem with Spot.Us’s crowdfunding mechanism is that it leaves open the possibility for adversarial organizations to slow down the publication of stories by purchasing a majority stake, as well as the possibility of Spot.Us to censor or heavily edit content that does not align with its interests.

Problems with Crowdfunding Research Grants

Currently, academic researchers can crowdfund their research by using one of the existing crowdfunding websites, such as Kickstarter, PledgeBank, or Spot.Us., but these crowdfunding mechanisms do not adequately address two general problems with crowdfunding research. The first general problem with crowdfunding research is that crowdfunding requires the crowd to have money to spend on research, but many poor and low-income people lack money to spend on research – not to mention access to the requisite computer technology, and so would be largely excluded from participating in crowdfunding. If the research agenda were significantly influenced by crowdfunding, poor and low-income people would be significantly excluded from influencing the research agenda. Crowdfunding, along these lines, would largely follow a research agenda biased for the research interests of those who already wield significant economic power.

The second general problem with crowdfunding research grants is that most people lack the expertise necessary to determine the value of certain forms of research. If non-experts cannot determine the value of certain forms of research, then they may fund research that yields relatively little value. This information asymmetry may not only lead to the inefficient funding of research, but it may also undermine the motivation of non-experts to fund certain forms of research. For example, if a person does not feel qualified to make certain funding decisions, then they may choose not to participate in making those and other funding decisions. In the extreme, non-experts, even those with money to burn, may decide not to participate at all in making any funding decisions, and so avoid crowdfunding research altogether.

Mechanisms for Crowdfunding Research Grants

To address these two general problems with crowdfunding research grants, we must redesign the crowdfunding mechanism itself. To allow more poor and low-income people to participate in crowdfunding, we must design a crowdfunding mechanism that provides poor and low-income people with money to spend on research grants. Since poor and low-income people lack this money, it must be gathered from other sources and redistributed to them.

The crowdfunding mechanism could allow government agencies, charitable organizations, and individuals to grant groups of people, including poor and low-income people, money to individually spend on research grants. To do this, the mechanism would need a way to allow people to create and join custom groups, and to be automatically included in demographic groups based on verified personal data. For example, the mechanism could allow an organization like the United Way to donate and $500,000 to the group of people who make less than $25,000 a year and who live within 50 miles of New Orleans, so that every person in this demographic would have some money to spend on research grants aimed at improving disaster relief for poor and low-income people. After the money was donated to the group as whole, the crowdfunding mechanism could divide the money evenly amongst all active members who’s profiles fit that demographic description. In this way, the crowdfunding mechanism would facilitate meta-grants, grants given to a group of people in order for its members to personally contribute money to other grants. By augmenting crowdfunding mechanisms with meta-grants, people who are directly affected by serious socio-economic problems, and who are currently excluded from participating in crowdfunding, would be able to directly make important decisions about how to spend scarce research monies.

To accurately distribute money to members of specific groups, the crowdfunding mechanism would also need a way to verify whether a user is a member of a specific demographic or user-generated group. For user-generated groups, authenticating membership is easy; users are members of all public groups they join, and administrators of private groups determine whether an applicant is a member. Determining group membership for demographic groups is more difficult. To determine whether a user is a member of a demographic group, the user must provide some personal information, and trusted third parties must infer demographic data from this personal information. For example, perhaps a trusted third party can verify the address of an applicant and estimate the income level of the member based on that address.

The crowdfunding mechanism would also need a way to redistribute monies if members become inactive, leave a group, or leave the crowdfunding service. This would prevent group monies from clogging up in the accounts of inactive users. One option would be to have a deadline for spending money contributed by a group, after which the money would be returned to the group account and redistributed to active members. If a user leaves a group or the crowdfunding service, their money would be automatically returned to the group account and redistributed to active members.

It is important that group money can only be spent on legitimate research grants, and not laundered through fake research grants, or through grants that give kickbacks to funders. To do this, crowdfunding mechanisms need to vet who can receive group money. Since the crowdfunding website cannot afford to do this by itself, it can use affiliated and trusted research organizations. For example, a university, college, library, museum, or archive could be a trusted research organization, and the crowdfunding website could allow those organizations to authenticate the researcher status of fundraiser. The research organization could confirm some minimal personal information of the researcher, including their name and address. To receive sure how to verify income level. Perhaps the verified home address of a member could be used to estimate income based on nearby property values.

To address the second general problem with crowdfunding research, we must design a crowdfunding mechanism that allows non-experts to delegate funding decisions to those they perceive as more expert. One way to do this is to allow members to entrust some of their money to other members that they perceived as having more expertise than themselves. This approach recognizes that a funder may not know much about a particular area of research, but they probably know someone who knows more about that themselves. If the crowdfunding mechanism allows members to delegate spending decisions to those they believe are more informed than themselves, then we can expect more informed and more efficient allocations of funding. For example, you may not be a physicist, but you may know a person who is good at math, so you may entrust them with $5 to decide on a physics research grant to fund. That person may in turn not know a physics major that they trust, so she may entrust the original $5 entrusted to her, as well as $10 of her own money to her physics major friend to decide on a physics research grant to fund.

So with this design, locally perceived experts can in turn entrust money to people they perceive as more expert than themselves. This process of entrusting money from those with less expertise to those with more expertise, continues from the local level and culminates in money that is entrusted with global experts. This mechanism of delegating money from non-experts to experts would allow non-experts to retain control of the money flow, but leverage the expertise of trusted experts. To do this, the crowdfunding mechanism needs a way for members to limit how their “entrusted” money can be spent by trustees. If those entrusted users become inactive, the money must be automatically returned to the person who entrusted it. As we can see each unit of money could have a trail of trust, starting from the original owner, who is the least expert, to the people he first entrusted his money with (the local experts), and finally the people he does not directly know, but that those he knows ultimately trust as most informed to make a funding decision (the global experts). This delegative model for grant monies leverages information assymetries to make more efficient funding decisions; it uses people’s knowledge that other people know more than they do to intelligently route the money from non-experts to local experts to global experts.

Republican House Whip Eric Cantor recently launched an online political campaign, YouCut,
for people to vote on federal budget cuts. According to an interview with Fox News anchor, Gretchen Van Susteren, Cantor promises to propose the budget cut with the most votes for “an up-or-down vote on the House floor the following week.” Cantor’s YouCut website claims that YouCut “allows you to vote, both online and on your cell phone, on spending cuts that you want to see the House enact.” If this is true, then YouCut would be an enormous advancement for participatory budgeting and direct democracy in America. But it’s not true because while YouCut is somewhat participatory, it is not democratic. YouCut is not designed for you to make decisions about what to cut; it’s designed for Eric Cantor and the Republican Party elite to decide what to cut. In other words, YouCut is really TheyCut.

There are two main reasons why YouCut is really TheyCut. Firstly, YouCut does not allow you to propose budget cuts. The agenda is set by Eric Cantor and the Republican Party elite. This means that the majority of Americans – including ordinary Republicans, Independents, Democrats, and others – cannot directly propose budget cut items for the YouCut community to vote on. Instead, budget cut items are privately culled, filtered, and vetted by Cantor and the Republican Party elite, and then put on the website. While many ordinary people may agree with Cantor’s selection of budget cut items on the ballot, the process systematically prevents the inclusion of budget cut items that go against the interests of Cantor and the Republican Party elite. Ordinary Republicans, Independents, Democrats, and others may approve of prudent budget cuts that conflict with the political and economic interests of Eric Cantor and the Republican elite. If Eric Cantor and the Republican elite censor these budget items from the ballot, we would never know it because there is no publicly available data to show rejected proposals. Clearly, Cantor’s YouCut has the potential to provide a very distorted image of American budget cut preferences.

The second main reason why YouCut is really TheyCut is that YouCut is vague and not legally binding. Cantor’s proposals are vague. They do not include the specific language he would propose as legislation. Moreover, even though Cantor promises to propose legislation that addresses a democratically selected budget cut, he is not legally obligated to do so. This wiggle room is afforded by the vagueness of the proposed budget cut item. While he suggests otherwise, Cantor has not entered into a contract with the American people to propose specific legislation because, in part, he is not asking people to vote on specific legislation.

There are other systematic problems with YouCut’s seemingly democratic design. It does not allow people to attach comments to the proposed budget cut items, thereby silencing useful criticism. It is also hosted on a seemingly Republican controlled website, thereby discouraging input from non-Republicans, and potentially biasing the sample of voters to Republican voters.

All of my criticism of YouCut may be written off as a misunderstanding of partisan politics – that I am expecting too much direct democracy from the Republican House Minority Whip. He is a professional partisan politician after all, and one who directly benefits from the status quo of indirect democracy. Some might ask why I am so surprised to discover another partisan ploy to rally political support with populist rhetoric, even sophisticated populist rhetoric, like the use of online voting that pretends to give political power to ordinary people about how to balance the federal budget.

The problem with this line of criticism is not that it assumes a tendency for political corruption, but that it accepts it without considering creative alternatives, alternatives like allowing the public to propose and vote on all proposed budget cut items, or allowing the public to post comments on any proposed budget cut item. These creative alternatives are not that difficult to imagine or implement, but they do move us closer to systematically rejecting political corruption in the legislative process.

On YouCut, Eric Cantor and the Republican elite ask us to “vote on this page today for your priorities and together we can begin to change Washington’s culture of spending into a culture of savings.” I am asking us to seriously re-examine whether YouCut actually allows us to democratically decide on our budget priorities, or whether YouCut allows Eric Cantor and the Republican elite to use us to provide a veneer of democracy for their budget priorities.

Eric Cantor and the Republican Party elite are not only generating media buzz and populist excitement about their campaign, but they also get your email address or phone number, depending on how you vote. You may be thinking that you are merely voting for a budget cut, that government is finally listening to you. Actually, you are adding yourself to a political list that might be used for other partisan purposes.

While YouCut is not designed for democratic decision-making, it demonstrates how easy it would be to design a democratic decision-making website, one that gathers votes from the public and connects those votes to legislative action. Ironically, while YouCut may be designed to manipulate the public for private interests, it gives us reason to think that we could design a website that actually overcomes those private interests by democratically voting on public interests. Currently, YouCut is really TheyCut, but in the not so distant future, we could really have a YouCut.