AGP discussion: Aragon Network Budget

I don’t think it was ever the case that AGPs are processed in some specific order and become effective in a current vote. I think if this passes it would become active for future ANVs only. Is it correct @light?

I’m a bit concerned. We are clearly pointing out that ANT holders cannot make reasonable decisions, and instead of focusing on that deeper problem we are already trying to quick fix it top-down. This sounds like implementing parental controls to those who make important decisions in the network :thinking:. Why not! But not exactly my ideal scenario.

Not sure that’s what you were refering to, but I think we could do better at providing data to inform this discussion in depth.

We haven’t worked yet on detailing different treasury scenarios to the rest of the community. I think that in order to have this budget conversation it would be very useful to have a google sheet with treasury forecasts:

  • Using current burn rate and growth rate over quarters
  • Splitting that between different grant programs and entities
  • Detailing scenarios with different ETH values, number of Flock teams… etc…

Kind of saying that to myself here as producing this kind of document could fall into AA’s scope.


I thought that was a placeholder, looking at the 1M DAI figure

My understanding is that the proposals is a “20 percent of treasury value OR 1M, whichever is greater”, or in the case where the budget period is quarterly “5% percent of treasury or 250K, whichever is greater”. Looking at the treasury value is 42M+, so that would work out to 2.1M “this quarter” if the budget policy was evaluated right now.

I think the exact numbers and mechanism should be discussed more, but it seems like a reasonable enough place to start the discussion.

IMO, implementing a budget is a tool that is helpful for prudent people, organizations, and groups to make better financial decisions.

I think presenting ANT holders with a budget, and having ANT holders make the decision to ratify that proposal is a good indicator that ANT holders are capable of making sound fiscal decisions. These are not “parental controls” imposed externally, this would just be the community deciding to adopt a more mature process for capital allocation.

1 Like

Correct, that is the intention.

+1 yes this is not a constraint coming from an external “parental” authority, but from the community itself. As I framed it in my opening post this is intended to be a “Ulysses pact”, a self-binding mechanism the community uses to recognize the unsustainable spending rate and do something about it. We cannot continue saying “yes” to every good looking proposal. A budget provides us a common agreement we can point at to justify why certain good proposals can be funded and others can’t.

More data on this is welcome, and will be useful for informing how, if, and when we make any adjustments to the proposed budget.

I would not say that ANT holders cannot make reasonable decisions. I would say that they have been well-intended but short-sighted with our treasury. But it is not too late to take stock, reflect, and change course. I am putting forth this proposal to give ANT holders an opportunity to do just that.

1 Like

I have made some adjustments to the proposed language in the first post on this thread. Thank you @lkngtn @stefanobernardi and @LouisGrx for your feedback.

1 Like

There’s a really big problem with this conversation. ANT is the lifeblood of the Aragon community, yet we’re talking about everything in terms of DAI and US dollars. We’re talking about misaligned incentives, but we’re not giving ANT to people who contribute to the community. Honestly I’m not sure why we’re surprised that things are the way they are. People who want DAI are not the same people who want ANT.

There’s a really really easy way for us to reduce our burn rate and align incentives for all parties involved. Right now we pay out mostly DAI and a little ANT as a success reward. We have X DAI and Y ANT sitting in our vaults. We could just switch that around and pay out mostly ANT and a little DAI. Funding secured! lol

But seriously though, ANT aligns incentives and gives people a voice in ANVs. Every other project bootstraps with their native token. The only ones who don’t are the ones that aren’t trying to build an actual community. Places like GitCoin have a lot of DAI bounties because the target market is temp contractors. DAOs and cryptoeconomic mechanisms are all about aligning incentives. DAI doesn’t align incentives for contributors to “feel like founders.” If we want want people to “feel like founders” we need to give them ANT.

This would put downward sell pressure on ANT as teams cash out to pay expenses. This pain is amplified by the fact that ANT has very low liquidity. If more people had ANT there would be more liquidity. Avoiding this problem and kicking the can down the road just sets us up for more nad more pain later on. We need to address this as soon as possible. ANT is the lifeblood of the Aragon network, and we need to drive value to ANT, but we also need ANT to be liquid. If we give contributors ANT then we’re both increasing liquidity and aligning incentives. This will allow us to focus on driving value to ANT and to have that value continue to support the community long after the initial USD treasury funds.

We need to be give contributors more ANT and less DAI. The budget needs to be an ANT based budget. If we drive value to ANT the budget will be worth more and allow us to do more. If we do not drive value to ANT, then the budget shrinks. This aligns incentives.


I hear what you are saying, but I think that is a separate discussion. This proposal denominates the budget in DAI because DAI is a relatively stable asset. The decision is for pragmatic measurement purposes only. Respectfully, I consider the specific token that recipients of funding choose to get paid in to be outside the scope of this proposal.

Measuring a budget in DAI/USD implies that ANT is not valuable. This implies that we want people to earn fiat from Aragon’s treasury vs working as a team with incentive aligned “equity type” compensation. The whole point of compensating contributors with ANT is that it is not a stable asset. It’s value fluctuates, and that incentivizes ANT holders to drive value to ANT. A stable asset does not align incentives!

This is the most crucial decision of any project’s budgeting. It’s the foundation everything else is built on top of. We need to denominate our budget in ANT. We need to calculate our burn rate in ANT. Our ANV/annual budget needs to be based on the ANT burn rate. Considering that the title of this thread is “Annual budget for the Aragon Network,” I think this is incredibly on topic.

Measuring the budget in DAI does not imply anything other than that DAI is a stable unit of account by which to measure our budget and spending. By that measure, yes ANT is inferior and probably always will be. ANT is good for aligning long-term interests in the Aragon Network, particularly when coupled with vesting. And it will soon be a useful tool for onboarding jurors into the Aragon Court. But ANT is not useful for measuring spending day-to-day. It is a utility or work token, not a medium-of-exchange token. For this reason, I do not plan to modify this proposal to denominate the budget in ANT. Thank you for your feedback.

Choosing to denominate a budget in a liquid, stable asset as reference value seems practical. Its worth noting that the budget does scale based on asset value held in the treasury so even though it is denominated in DAI, the budget will increase if asset value (DAI or ANT or whatever else is held) increases in value.

I think it would be interesting to see a proposal or simply a shift in norms such that ANT is used and distributed from the AGP process, but definitely think that it should not be bundled with the idea of creating a budget, as there are lots of concerns unrelated to the idea of implementing a budget that would involve.

1 Like

The new section related to the Aragon Network vote ballot and example seem clear to me.

However, the first section about total amount funded by Finance track AGPs and example seems a bit confusing, as the examples still seems to be based on a yearly process but the explanation has been updated to reflect a quarterly budget period.

I’m also not certain we need both sections, perhaps simply having the second section and indicating that proposals that by themselves would be over budget should be rejected before being put to a vote since it would not be possible for them to pass.

It seems that this first part is “pre-approval”, it is asking the association to reject (prior to voting) any proposals that would exceed the budget by themselves.

I would like to clarify:

  1. If the budget period is quarterly, is this specifically to indicate that the Association should reject any proposals that would exceed the quarterly budget by themselves? If so I think we may only need one example, and the example should probably be updated to reflect the quarterly budget period amounts.

  2. If this indicates that there is both a quarterly and year-to-date budget policy that is being evaluated, then shouldn’t the year-to-date policy be described in the explanation before the examples?

@anteater can you clarify your intention here?

I made a hasty edit to the first post on the thread and forgot to remove the examples pertaining to the first iteration of this proposal. These have been removed. Thank you for pointing out this error.

The intention now is to have quarterly budgets, and have Finance proposals during each ANV assessed against the next quarter’s budget (the quarter immediately following the conclusion of the vote). The way it is worded right now, the assessment as to whether or not a proposal is within the budget occurs after the proposal has been approved. This is because we do not know how to rank proposals against each other until we know how ANT holders have voted.

To visualize:

  1. Everything works as it does today up until an ANV concludes
  2. The Association sees which Finance proposals were approved and what the absolute support was for each proposal
  3. The Association measures the value of the treasury, calculates the current quarterly budget, and tries to process the proposals that were approved in descending order by absolute support
  4. The Association pays out the proposals that fall within the budget, in that descending order.

One thing that might need to change in this version is when we calculate the treasury value. Since most of the treasury is in volatile ETH and ANT, a lot could happen between the date proposals are finalized and the date that the vote ends and proposals are processed. To reduce uncertainty for proposal authors, we might consider changing the assessment date to be the day proposals are finalized (or even earlier, to give a chance for proposals to be adjusted as needed). Am open to suggestions about this. I do not think there is a “right” answer. I chose to do it the way it is now since measuring the budget at the time proposals are processed is the most accurate way to determine if they fit in that budget. But maybe we can find a solution that is most proposal author friendly and budget friendly.

1 Like

Hey again @anteater! Thank you for bringing up this AGP idea. I generally support the idea of a Network-wide budget, for many of the same reasons you and others mention in this thread and elsewhere. I have some questions about implementation. Specifically:

  • How will Finance track proposals with recurring payments be treated? For example, the CFDAO is currently funded quarterly on a recurring basis, essentially until either the project runs out of money or ANT holders modify the payment schedule. Will proposals like that have to be re-approved each quarter/ ANV under this proposal? Or will they continue on auto-pilot so long as they fit within the budget?

  • Following on that, in the proposal ranking scheme described, where would recurring proposals that have been approved in a previous vote fall? Would they try to get “processed” first, last, or some other way?

  • Similarly, is the expectation that big funding proposals like Flock or the AA budget would not longer be made on an annual basis, but on a quarterly basis?

I hope these questions are clear! Cheers

Edit: I think this quote below answers my last question, so to be sure, you’re saying no one could request funding intended to last beyond the current(/next) quarter? Or could one team take up the whole budget allocated in one ANV cycle and use that budget as their annual funding, then another team could create a proposal to use up the entire budget in the next ANV cycle for their annual funding, and so on?

I think this is a good suggestion and addresses the “front-loading” issue. At the cost of a larger number of Finance track proposals to review throughout the year (four from each Flock team, minimum) but seems worth it.

Good questions. Recurring payments are not explicitly addressed in the proposal. But I intend to disallow recurring payments. A Finance track AGP can only spend from the next cycle’s budget, it cannot spend money from a future budget cycle beyond that.

Implementation-wise, how to treat recurring payments, would be to write into the AGP description itself that all Finance track AGPs with one-time payments that have already been approved at the time this AGP is approved will be paid in full. All previously approved AGPs with recurring payments will continue until they either expire or two more ANV cycles are completed, whichever comes first. ANT holders will then have two ANV cycles before the payments stop to decide whether to continue the payments on a quarterly basis under the new budget, or not.

Recurring transfers are useful because they allow the default behavior to be to continue to fund an initiative until the recurring payment is updated, and doesn’t require voting on the issue every quarter–but also doesn’t require approving a lump sum payment all at once. If a recurring payment is competing for space within a budget, there is always the opportunity to update the amount or cancel the payment during an ANV cycle…

My suggestion would be to adopt the following process so that we can continue to support recurring transfers but ensure they can be canceled before impacting an ANV budget:

  1. First process proposals which reduce or eliminate a previously approved recurring transfers.
  2. Debit and currently recurring transfers from the ANV budget.
  3. Then Process proposals which make new transfers or increase previously approved recurring transfers using the process described above.

I don’t think there is any reason a proposal couldn’t provide funding for something intended to be spent over a longer period of time than a quarter, just that the network cannot exceed the budget within a given quarter.

My expectation is that the result will be more collaborative budgeting and planning as proposers try and coordinate across organizations so that their burn rates and proposal submission align together rather than compete for limited space.

1 Like

This brings me to what I was getting at with my questions before. It seems like allowing recurring payments then would kind of defeat the purpose of doing this quarterly instead of annually, wouldn’t it? In that case, my strategy as a “high spender” would be to create one or more payments at the beginning of the year that repeats quarterly and uses up as much of the available budget as possible. Or several “high spenders”, Flock teams for example, will all create their annual budget AGPs in the first quarter of the year and say “this budget repeats quarterly until the end of the year”. Which has the same effect as making the budget annual and not quarterly. So all available funds could effectively get allocated at the beginning of the year, then AGP authors only have to possibly do another AGP mid-year if they need to lower their proposal amount to fit into the budget based on current ETH / ANT prices.

With this in mind (and happy to get any pushback on these concerns as well) do you still see an advantage to making this a quarterly budget instead of annual?

I don’t think so, recurring payments can be altered throughout the course of subsequent ANV before funds are allocated. In contrast a yearly budget would allow the entire budget to be exhausted in a single ANV.

I don’t really see an issue with this, it may even be desirable as it would encourage longer term planning and coordination.

I think the key thing is that a recurring “budgeted” allocation can be altered throughout the year, making room for new information and new potentially higher leverage proposals to emerge and replace previously scheduled transfers.

From a strategy perspective I as a “high spender” I would probably not want to create a single proposal/recurring or otherwise that takes a large portion of the quarterly budget because the more a proposal has the potential to crowd out other proposals the heavier the competition is for that part of the budget. I can reduce risk by submitting multiple proposals for different strategic initiatives because each can be approved or declined independently. I may end up with a very high likelihood of receiving 70% of my requested budget versus an uncertain outcome between 100% and 0%.

1 Like

This doesn’t feel much different than an AGP that “cancels” a previously approved year-long Flock proposal. But I don’t have a strong opinion about this. I’m mainly interested in passing a budget and see how that affects the discussion about spending prioritization. So I’ll support quarterly budgets, and allowing recurring payments (I also like that we don’t have to renew AGP-10 every quarter).

I’m not really clear on whether it is possible (legally) for the association to reclaim funds from a funded flock proposal based on an ANT holder vote. I don’t really have visibility into that process, but generally even if it is possible I would not be supportive of any such proposal because I don’t think that expectation was clearly communicated.

My perception to this point is that, much like on-chain transactions, we should assume that transfers to flock teams irreversible, and if the intention is to approve a transfers on the basis of hitting milestones or funds to be released over time we should have the process actually transparently structured that way such that transfers occur over budget periods or conditionally upon milestone achievements.

@anteater as you are championing this proposal, do you still intend to submit a proposal that removes the ability to create recurring finance track proposals?

True, these are important considerations. There isn’t an explicit ban on such an AGP. I was considering a case where an approved Finance track AGP became very undesirable and the AA hadn’t paid off the whole thing yet (for whatever reason) and ANT holders decided to pass an AGP blocking additional payments related to the old AGP. It’s an extreme case but, if not allowed, it is at least not explicitly banned in AGP-1. Maybe that could be clarified in an AGP (this one, if it is ever submitted, or another).

1 Like