Draft AGP for Vote #2: Futarchy Signaling Markets

See Pull Request 30 on AGP repository, posted by Level K

“Aragon is a fight for freedom. Aragon empowers freedom by creating liberating tools that leverage decentralized technologies.” - AGP-0, The Aragon Manifesto

Futarchy, described as “Vote your Values, Bet Your Beliefs,” offers a collusion resistant way for organizations to make decisions. By using prediction markets, participants are incentivized to evaluate a decision based on a given metric. For organizations aligned around given values, futarchy offers the promise of enabling well informed decisions without delegating power to the few. Futarchy offers the potential to be a liberating tool, and advancements in decentralized technology are allowing it to come to fruition. We seek to demonstrate and improve upon an implementation of futarchy, better understand futarchy in practice, and to educate the Aragon network and the public at large on the promise of Futarchy.

Our team has been developing an Aragon app for futarchy decision markets, as part of this Nest proposal. As the first use case for the futarchy app, we’re planning to build “Signaling Markets” for AGP votes using ANT price as the metric.

The AGP Signaling Market will allow users to make bets on the price of ANT at a future date, given the outcome of each AGP decision.

Example: “Team X” requests Flock funding. Two markets will be created - 1) “If Team X receives funding, what will ANT price be 6 months from today?” and 2) “If Team X does not receive funding, what will ANT price be 6 months from today?”. Users can take “Long” or “Short” positions in either of these markets. The desired outcome is to predict which decision (“Yes” fund Team X, or “No” do not fund Team X) is best for ANT price.

The futarchy app provides the smart contract infrastructure for setting up these markets, as well as the UI for users to buy and sell positions.

It’s important to note that these markets are set up for “signaling” the impact of each decision on ANT price, and will not have an effect on the actual AGP decisions. ANT holders can look to the signaling markets as an input for their voting decisions. Smart contract based voting delegation to market decisions could also be built, although this is not in scope for this proposal.

Level K is requesting $30,000 of funding for:

  • Designing and implementing signaling market specific UX/UI, to launch an instance of the futarchy app for AGP’s.
  • Modify futarchy smart contracts to be signaling markets instead of decision making
  • Add Tutorial video(s) and informational material on how to interact with the signaling markets
  • Build a script or CLI that allows us to create new markets for each new AGP
  • Build an oracle that resolves based on AGP proposal decision vote outcome
  • Build an oracle that resolves based on ANT price, leveraging output from the Oracle Manager
  • Provide some initial liquidity for the markets and cover deployment costs
8 Likes

The proposal scope is extremely clear, complete and interested. I like especially that you propose to create educational material on such a complex topic, kudos for that!

I think the markets or any kind of measuring for the decisions that affect the market are needed - brilliant usecase.

I have a question, however (edited after reading @maria’s feedback on your nest proposal) :
Re futarchy, i would really like to understand more of an Aragon-wide alignment with this particular form of governance - not for clarifying on this proposal, but in general.

2 Likes

This is brilliant. However I have one question: how do isolate votes in a ballot, in a way that you can attribute the metric changing because of a given vote, and not the whole ballot?

BTW, small thing: Nest is not an acronym (e.g. not NEST)

2 Likes

Hey @johnkelleher ,

Looking forward to seeing implementations of the Futarchy app :)!

I have concerns regarding using ANT as the metric. In your example - If I’m not missing the point - ANT’s price is supposed to rationally reflect the work of team X after a 6 month period. Given the state of the market, I’m afraid that in the short term we cannot 100% rely on ANTs price to reflect any real-world events.

Imo, in this case, a product related metric (active users for such or such feature, nb of DAOs created, user adoption etc…) could be better proxys until ANT markets become more efficient/rational. What do you think?

3 Likes

This is brilliant. However I have one question: how do isolate votes in a ballot, in a way that you can attribute the metric changing because of a given vote, and not the whole ballot?

Great question! There is the possibility that multiple futarchy markets using the same metric could be confounding. We are hoping these signaling markets and other pilots of futarchy will help us gain an understanding of these dynamics.

The work being done by Alan Lu and others at Gnosis with the Gnosis Prediction Markets 2.0 could also increase the design space for futarchy and allow people to participate in more complex conditional outcomes. Level K is hoping to upgrade the futarchy implementation to these contracts soon!

3 Likes

Love this idea! We think it would be great to explore different metrics, and we could possibly launch 2 or 3 different types under the efforts described in this proposal.

Something to consider is how easily the metric could be manipulated. We would be interested in hearing more about your product related metric ideas and to think a bit about which ones might be best suited for these futarchy signaling markets.

3 Likes

@LouisGrx Mike from Level K here. We’ve also thought about using a combined “health score” metric, which could be an average of ANT price and other metrics like the ones you suggested. Maybe the inputs of a such a health score could be voted on by ANT holders.

For the initial launch, like @johnkelleher mentioned, we would be interested in setting up multiple markets with different metrics, in order to analyze them separately. We want to be careful about introducing manipulation incentives (and also watch closely for evidence of manipulation). Number of DAO’s created, for example, might be easy to manipulate - take a large LONG position, then create thousands of DAO’s before the market resolves.

Getting metric information on-chain is also a challenge. “Number of unique new users” for example isn’t something that’s readily available for us to use as an oracle. We have a Nest proposal to create an Aragon app that could provide a decentralized way to bring these metrics on-chain.

6 Likes

I like the “health score” idea as an aggregation of multiple metrics.

2 Likes

As someone who newly got involved in the community, I recognize the value in something like this.
To engage in Aragon’s governance, one needs ANT.
I have been acquiring some over the course of the past weeks, but ideally I would prefer the system to value my governance contributions and reward me accordingly.
Then I could buy a modest amount of ANT, bet on the right choices, and earn more over time. As time goes on, it compounds, and I would end up having a sizable amount of ANT tokens.

4 Likes

Our team has been developing an Aragon app for futarchy decision markets , as part of this Nest proposal . As the first use case for the futarchy app, we’re planning to build “Signaling Markets” for AGP votes using ANT price as the metric.

It’s important to note that these markets are set up for “signaling” the impact of each decision on ANT price, and will not have an effect on the actual AGP decisions. ANT holders can look to the signaling markets as an input for their voting decisions. Smart contract based voting delegation to market decisions could also be built, although this is not in scope for this proposal.

Yay for signalling! We need more of that :slight_smile:

Edit: it’s recently come to my attention that a signalling app for Aragon already exists. Would this proposal integrate into that, and if not, in what situations would one be better than the other?

I have concerns regarding using ANT as the metric. In your example - If I’m not missing the point - ANT’s price is supposed to rationally reflect the work of team X after a 6 month period. Given the state of the market, I’m afraid that in the short term we cannot 100% rely on ANTs price to reflect any real-world events.

Imo, in this case, a product related metric (active users for such or such feature, nb of DAOs created, user adoption etc…) could be better proxys until ANT markets become more efficient/rational. What do you think?

I agree 100%. The price of ANT is determined by Mr. Market and his drunk cousin Mr. Cryptocurrency Market. They’re both very irrational! From what I’ve seen the price is more often determined by the price than any underlying value lol

Also, in the case of introducing multiple markets, those markets will only work if there is enough engagement. If the already scarce attention is dispersed into multiple markets/channels/things, it could actually create more noise than signal because those markets would have low volume and be easily swayed. Maybe it would be worth finding 1 or 2 concrete metrics to start with and then go from there?

4 Likes

What is the relationship between the futarchy app and the futarchy courts?

Love this! I had a chance to play with the futarchy app recently and I was very impressed. I would also echo the concerns about using ANT price as a metric, I don’t really think there is much correlation yet. however it is just signaling and still a valuable experiment, id like to see it in action.

do you guys have some rough estimation on how deep the liquidity needs to be for each market and what factors need to be taken into account

They are currently being developed separately, the court has a simple final appeal mechanism that results in all jurors being asked to vote (so simple majority of jurors), but eventually if the Aragon Network decides to go the futarchy court route this could be appealed to a futarchy decision market.

So there would only be one futarchy app implementation, the Aragon Network would just use that as the final arbitrator for appeals. There is no intention currently to have two separate implementations of futarchy.

Does that help clarify?