Common Good

Common Good is a blockchain-based application enabling the coordination and motivation of stakeholders in producing common goods.

Written with Latif Eliaz, Ido Gershtein, Shahar Oriel and Nimrod Talmon.

Introduction

A common good (CG) is a product or an outcome which benefits almost everyone in a certain community,¹ may it be a neighborhood, the Ethereum ecosystem, a nation state or entire civilization. As such, everyone in that certain circle has the interest to see the common good produced, while no one has the incentive to produce it on their own. The total benefit to all people in a community gained from a common good goes above its total cost, by the definition of a common good; but, its gain per individual or entity is much lower than the cost of its production. The inability to act on it is a mirror manifestation of the tragedy of the commons.²

Today, countries, municipalities and NGOs are the entities that supposedly take care of common goods, but their capacity to do so is very limited due to their centralized structure. They are limited by the relative ineffectiveness of centralized constructs — in sense-making, scalable action, engagement and alignment of interests, and more severely, by the personal interests of the people steering them, which often override their interest to take care for the benefit of the community they are in charge of steering. Indeed, neglecting such common goods is one of the biggest problems of humanity in almost every possible domain and circle we can think of.

In the future envisioned here, decentralized networks play the role of governments, municipalities and intentional commons, fostering common goods. It is possible to produce common goods when a big-enough community cooperates to bear the cost of production and its implementation; but this, correspondingly, requires large-scale coordination, and large-scale coordination is generally a very hard problem. In this article we introduce Common Good, a blockchain-based application that solves this problem by enabling the coordination and motivation of different relevant actors for achieving a desired common good, by providing it with a “business model” just as in the profit-seeking sector. Our solution takes inspiration from the Social Impact Bonds (SIB) model.

Prescription

A common good (CG) process begins with an initiator proposing the production of a common good. Then, during the predefined lifetime of the process, funders who care about this common good may pledge funds for its production, being reassured that their money will only be used retroactively, had the common good been eventually produced — no risk taken. Executors who wish to produce the common good may do so, being reassured that they will be compensated by the pledged funds had they been successful. And profit-seeking investors may buy a portion of the potential reward from executors (in the form of per-executor tokens that are made redeemable against the future reward had they been successful), and by that provide them with liquid funding for operation. Finally, if and when executors achieve the desired outcome, as decreed by a predefined judge, the pledged funds are released as a reward to the successful executors and the investors who bought their tokens. If no success has been reached after some predefined limit of time, the funds go back to the funders who provided them. Executors and investors only see profit, and funders only spend it, if and only when the common good is produced.

In the simpler case of competing executors, only a single executor is declared a winner, who then wins the entire pledged fund. In the more complicated case of cooperating executors, multiple contributing executors share the reward according to the relative contribution made to the completion of the final result. In this latter version, an additional evaluation process is introduced to evaluate and decide about the relative value contribution made by each executor to the final outcome;³ and accordingly an evaluator, also defined within the setup process. In both cases, investors are motivated to scout for the most promising projects, and might also be encouraged to increase fundings to them in order to guarantee the success of their chosen projects with respect to the other ones. Binary success (“all or none”) is used in this article, although a continuous-success model is also valid and will be more relevant in some cases.

We’ll now describe the entire process in more detail, step by step.

Initiation

The initiator is setting up the CG process and providing the following info:

1. CG process target: The definition of achievement of the proposed common good, preferably via a set of objectively measurable defining conditions (e.g. KPI’s),⁴ and the total time frame given for producing the CG. If the CG is not declared successful after this period, pledges are released back to the funders.⁵
2. Raising target: The minimal total amount of pledge that needs to be collected in order to continue the CG process, and the maximal time frame to collect it. If the raising target is not met within that period, pledged funds are released back to the funders.
3. Judge identity: the address of the judge, which could be an individual, a multi-sig account of reputable actors, a decentralized court system, a DAO or anything else chosen.
4. In the more advanced version of cooperative executors, an evaluator identity needs to be provided as well.

Funding

Funders who care about this CG may now pledge their funds to the CG process. The funds will be locked in a smart contract until either a success—of the entire process or a milestone — has been achieved, or the time limit allocated for it has passed.

Execution

Actors may now sign up as executors. When signing up, the CG process mints each executor their own token supply (say, 1M tokens).⁶ These tokens will be redeemable against the entire fund—in the case of competing executors, or against a portion of it—in the case of cooperating executors and pro rata to their “contribution score” as determined by the evaluator, once and if the CG is successfully completed as declared by the judge.

Once executors have their own tokens, they may now sell them to investors to obtain liquid funding for their operation. Executors and investors will be able to redeem these tokens for the corresponding share of the funds allocated to the project, if successfully completed. In essence, the pledge made by the funders provides a “business model” for executing on the related common good. Given that business model, now executors can raise funds from investors.

Investment

Investors can now invest in the executors they are willing to bet on by buying a portion of their tokens. These tokens are made redeemable against the pledged fund upon a success. In the case of competing executors, a single winning executor wins the entire fund, and in the case of cooperating executors, they share the fund pro rata to each one’s “contribution score” as determined by the evaluator. In both cases the investors have the regular market incentive to identify successful executors and bet on them. Given the sunk costs of investment, investors may further be incentivized to keep supporting the projects in order to maximize the probability of recouping their investment.

As an example, in the case of a single winning executor, and if funders already pledged $1M to the CG process, then an investor may trade with some executor 10% of her tokens in exchange for, say, $70K, reflecting a belief of the investor that there is at least 70% chance of that executor to be eventually successful.

For simplicity, we currently leave the token-sale process outside of the current app (i.e. the executor’s sale of tokens to investors happens somewhere else), but we discuss it due to its importance to the general flow.

Judgement

Anyone can signal that the CG process has been completed, and then the judge needs to approve or reject it. Upon approval, the pledge is released to whoever holds the executor’s tokens. Upon a rejection the game goes on, until the time limit for the process is reached. While the judge can be a multi-sig account of reputable actors, it can also be a decentralized court system (like that of Kleros), or a DAO. The signaling of CG completion and the opening of the judging period should be made costly — since judging is costly, and is otherwise spammable. We ignore here the definition of the fee or stake needed to be included.

Evaluation

In the app version of cooperating agents, at the end of (and possibly also during) the CG process, an evaluation process needs to be included, in which the evaluator decides the relative value contributed by each executor to the final result. Each executor obtains her “contribution score” by the evaluator, which then fixes the portion of the reward gained by that executor, and the investors buying her tokens. The evaluator is likely a DAO, which could be internal — composed by executors and pledgers, or external. The evaluation can be made retrospectively, in a single snapshot if and when the goal has been achieved; or continuously throughout the process, with a monotonously increasing score distribution. The latter may give more confidence to investors in picking and investing in successful projects. Evaluation protocols of this kind is a rising topic but is still in its infancy, and we defer the discussion of the evaluator and the evaluation protocol to future work.

Comments

A few short comments only on the most burning issues.

The Perfect Judge

Clearly, one of the key challenges of this app is providing a “perfect” judge. The judge needs to be honest in the case of objective measurables, professionally competent in the case of subjective measurables, and always live, reacting towards a signal of success.⁷ While in the short term we believe that a multi-sig account of reputable stakeholders or a Kleros decentralized court system may suffice, in the long term a more adequate and robust judgement procedure should be developed.

The Free-Rider Problem

Funders pay for the resulting common good and they take no risk of “losing” their money without getting back the desired outcome. However, since the desired outcome is a common good, those in the community who do not provide funding benefit from it as well. On purely economic grounds, this may incentivize players not to provide funding, which is the well-known “free-rider problem”. We make three comments on this regard.

Firstly, we argue that there are enough common good examples which would have enough stakeholders agreeing to paying-for-success for them despite this problem, and in that sense it is a viable model as is. Enough people will agree to pay enough funding for cleaning up the oceans from plastic, making peace in the Middle East, developing new antibiotics, or developing layer-2 scaling solutions for Ethereum — despite others not paying for them as well.

Secondly, we would still like to see a “business model” for funders rewarding them for their pledge in common goods, in addition to the resulting outcome. For one example, imagine a global Common Good DAO, envisioning to become the biggest global economic power and manage the global future currency supply. This DAO may mint future currency for the production of common goods now, and it may provide enough of it to make it worthwhile for funders of common goods, who then become investors in this future. Once enough people believe in and stand behind this DAO, it may become a reality.

Lastly, it should be noted that the same model can also be used for funding non-common goods, goods which benefit only their funders. Say, funders who ask to produce a certain product which has not been produced yet and who are willing to pledge for it in advance, but only to be transacted if and when it has been produced. This is an advanced version of Kickstarter.

Retroactive Public Goods Funding

The Common Good model which we developed in the past few months is very similar to the recently introduced Retroactive Public Goods Funding model introduced by Vitalik and Karl. It would be great to understand the differences between the models if they exist and to combine efforts.

Examples

In the simplest case of a single executor (who is probably also the initiator), this app is a sort of an advanced version of Kickstarter or Patreon. An executor is offering to produce something in the world, may it be a common good or a real product. Funders provide funding for that executor, but only provided that he successfully completes his offering.

In the other simplest case of a single funder (who is probably also the initiator, and potentially part of the judge), this app is a sort of an advanced bounty program platform. Funders propose missions to be completed, whereas executors can propose to complete them and win the bounty upon completion.

The landscape of opportunities becomes much richer and more interesting in the general case of multiple cooperating executors (with multiple fractional winners) and funders, providing the funding and conditions to produce much larger missions only achievable by multiple agencies — missions which are rarely produced today.

The list of examples is literally infinite. There are many environmental examples varying from locally cleaning a beach or a river pollution to globally cleaning the oceans from plastics, purchasing and protecting Amazon lands or fighting global climate change.

It can be used to fund large-scale medical endeavors not ventured today (such as developing new antibiotics), solve the hardest open questions, say, in mathematics or engineering, support class-action lawsuits or global awareness campaigns, send commons-based missions to space and even form the basis for a new global virtual state.

There are also examples of ecosystem common goods such as Ethereum’s, funding the development of layer-2 scaling solutions, gas relay networks, security tools and infrastructures, advanced wallets and more. Similar examples are relevant in other software ecosystems, Web3 and non-Web3.

There are smaller examples such as artists who offer to produce certain art pieces, cooperative funding and pre-purchasing tickets for large events, a cooperative purchase of a stretch of land or an urban neighborhood, and the list goes on and on.

Future directions

There are many directions this app can be extended towards, and we mention very briefly just a few:

• From competition to cooperation: We already mentioned this one above, but we mention it again since likely the first version of this app will be launched only with a single winning executor (thus, competing executors), whereas the more interesting cooperating executors model will come next, requiring the non-trivial evaluation model incorporated.
• Collaborative executors: Cooperating executors means that they work independently but eventually share the reward. Collaborating executors means that executors also build on top of each other, requiring much more elaborated decentralized coordination tools.
• Judgement fee: As mentioned above, some sort of judgement fee or stake needs to be incorporated.
• Continuous (fractional) success: A success doesn’t have to be binary, it could also be defined in a continuous manner, thus rewarding executors in proportion to their continuous success.
• DeFi interest: Pledged funds can be invested into DeFi to gain interest.
• Grant NFTs and tokens to funders: The CG process may grant NFTs to funders upon the success of the process. More generally, the CG process may grant tokens to funders, thus making them also investors (distinguish now “primary investors” — buying those tokens, from “secondary investors” — investing in the reward provided by the primary investors).
• Nested CG processes: We can imagine a CG process within a CG process, where the “mother” CG DAO can also pledge its own tokens into milestone, sub-CG process.
• The Common Good DAO: the same nesting can also work upwardly, forming a DAO for global common good. This DAO may grant its tokens to successful CG processes, and by that providing a business model for the funders of common goods. This is a vision we would very much like to advance and collaborate over.
• Different economic models: there are different economic models, tying the pledge of funders to executors and investors in different ways, to play with.

Outlook

The Common Good process seems like a viable and appealing model for the funding of common goods. There are countless examples of common good use-cases which currently do not have a viable funding model, and which it will be exciting to experiment with. We noted a few examples above, but clearly we are just scratching the surface of what is possible.

Lastly, this project should definitely be a common-good project by itself, which would mean:

1. To develop the very minimal v0.1 of it, and then use it to fund and develop its further evolution.
2. Ensure the resulting platform to be open source and owned by the commons.

Would you like to collaborate on this project? Reach us out on Twitter, Telegram or DAOtalk.

We would like to thank Shay Zluf and Amir Nathan for providing ideas for this work, and to Primavera de Filippi for commenting on the draft of this article.

¹ To be more precise, the term “common good” is used for non-excludable goods (defined relative to a certain community), rivalrous (say, fish) or non-rivalrous (say, open-source code). The latter case of non-rivalrous, non-excludable goods is also called “public good”.

² “Tragedy of the commons” refers to the case where everyone acts in a way which optimizes their self interests, given that everyone else does the same; while everyone doing so is actually sub-optimal for everyone. The inability to act on a common good is the same with “acting” replaced by “not acting”.

³ In principle, this process can be ex-post or a continuous process, the latter reducing the investors risk.

⁴ The more objective those definitions are made, the more the judge’s role resembles an Oracle. The more subjective they are, the more it will resemble a professional committee.

⁵ In a more advanced version of this app, the final target would be generalized to multiple mid-term milestones, and correspondingly multiple time frames after each of which if the milestone has not been met the funds are released back to funders. The next information of “raising target” would then simply become a first possible milestone.

⁶ In a more advanced version of this app, the executor could also introduce her own token supply, which is then acknowledged by the CG process and made redeemable for the pledged funds upon a success.

⁷ Later, an additional judge stake can be introduced to incentivize the judge for a reaction.