Algorand Community Governance Period 1 Review

The Algorand Foundation is excited to share the results of the first Community Governance vote on the Algorand Ecosystem Resource Pool. This is a significant moment for the Algorand network as it marks the first permissionless, decentralized governance vote to take place on the Algorand network and moves Algorand even further on its path to a fully permissionless, decentralized blockchain. Further, this vote marks the largest community governance vote ever in blockchain history.

The Vote:

For the first community governance vote, the Algorand community voted on a choice of 2 options on how community governance rewards would be distributed in 2022. These options were A: a simplified commit and vote model with no slashing or stake penalties – or B: a higher annual reward pool but with 8% slashing for governors who did not maintain their commitment or failed to vote. Read here for more information.

The governor community who were able to vote ( our total electorate ) were those that committed to the first governance period via the Algorand Governance app and who had maintained their committed amount in their governance wallet. At the start of Governance Period 1 ( Oct 1st to Dec 31st 2021 ), this electorate numbered 70,927 governors. This group of governors committed 1.88Bn Algo to governance ( representing over 30% of all Algo in circulating supply and equivalent to more than $3 Billion USD ). These numbers, in terms of governors, committed tokens and fiat equivalent value, represent the largest governance electorate in the history of blockchain.

Voting opened on November 1st and ran for 14 days, with a following cool-off period that ended on November 19th 2021. At the end of this voting and cool-off period, over 51,700 governors, with 1.8Bn Algo committed, had voted. This represented ~78% “turn-out” of the Algorand electorate and displayed a high level of engagement in the voting process. The final results were:

Option A: Simplified Commit Model: 56.58% (1.02Bn Algo )

Option B: Higher Rewards w/ Slashing: 43.42% ( 0.780Bn Algo )

Option A was successful and Governance Rewards will continue with a simplified, no slashing mechanism in 2022, with a reward pool of 282M Algo.

Long Term Commitment:

Reflecting on the initial vote results, Massimo Morini, Chief Economist at Algorand Foundation, said: “Apart from the unprecedented success in terms of participation and interest, there’s another element I’d like to point out. The governors have chosen by themselves a lower amount of rewards for the next year, which also implies a higher amount of rewards available in the longer term. Since Decentralized Governance is a path that will bring more and more choices and services to be led by governors, the fact that Governors have decided to set aside an amount of resources for their rewards in the future, rather than receiving them now, can be a sign of more long term commitment. A bit like them saying: “No need of slashing to keep us committed. We are committed already, and we are interested in a commitment that will extend well beyond a 3-month governance period”.

Massimo continues “There can also be an element of aversion to uncertainty. For the Algorand network, rewards are a compensation for services to the network. Some players can still be wary of any rules that make the choice of withdrawing commitment from the service of governance harder. You know, our DeFi ecosystem is growing, and some Governors may at any moment decide to commit their resources to some of the nascent DeFi activities. The more decentralized governance and other decentralized economic activities will interact in the future, the less this point will be relevant.”

In closing Massimo notes “There are, in fact, several merits in the choice of the many governors that decided to vote for stronger commitment rules. In the long preparatory work for governance, we incorporated in our governance some voices of the community that advocated the need of a “cooling period” after the vote. If voters retreat from commitment in that period, their votes will not be counted to compute the final result. This is an incentive for governors to avoid opportunistic or superficial choices, followed by rapid retreat after voting. Stronger commitment rules such as locking or slashing would strongly reinforce this. The truth is, in any case: we are writing a new page in the history of community governance, economy and human collaboration.”

The Algorand Governors (Electorate Analysis)     

The team would like to thank Menno Sijben of http://algorandstats.com/ for providing the charts and tables that underpin the following analysis.

The following is an analysis of the makeup of the Algorand governors, our first electorate and how their number changed during the governance period, the vote period and the cool-down period. This analysis will be updated in January 2022 to reflect the governance rewards claimed as a result of this first Governance period.

At a high level, we saw the following trend through the period, with a decline in stake and significant reduction in governors occurring after 14th November. November 14th marked the end of the voting period and all those governors that did not vote, became ineligible. In excess of 80M Algo and almost 20,000 governors left governance at this point.

What happened here? Did a large number of governors ( with their associated stake ) decide to exit the program and what was the makeup of this group? We must first look in greater detail at the makeup of Governors at the end of the commit window for Governance Period 1 and what happened to this dynamic during and after the voting period. What we shall see is that, in fact, there were 2 dynamics occurring simultaneously. One dynamic for the number of governors and a separate, and we shall see, specific event in terms of stake.

As you can see from the charts above ( eligible Governors on October 14th ), there was a very high percentage of governors with a stake of 1 Algo or less. When we look at the chart of the period after the close of voting, we can see that there has been a significant reduction in these 2 groups around the 1 Algo stake.

And the bulk of governors who became in-eligible also came from this group:

From the size of the av. holdings in this group ( circa 1 Algo ) it might be suggested that there was a large number of users who were cautious in their engagement with the governance tool and first tested their use of the governor commit process by making a ~1 Algo test governance commit. Upon successfully using the application, these governors would then have made a more substantial commitment, from a different wallet dedicated to governance. These test accounts then remained eligible until the end of the vote period and when no vote was made, these accounts became ineligible.

While this analysis explains the reduction in governors at the end of the voting period, it does not explain the >80M Algo reduction in eligible commit. When we look at eligible commitment by size at the end of the voting period, we see a similar chart to that at the start of the Governance period.

However, when we look at ineligible commitment by size we see a significant level of ineligibility in the highest commitment band.

If we look in detail at the ineligible commitment addresses in detail we see what has occurred. One wallet, with a commitment of ~80M Algo became ineligible.

On review, this wallet/account had a single transaction occur which reduced the Algo in the wallet below the committed amount. Since this is one of the fundamental criteria of eligibility – to maintain a wallet balance above the governance commitment at all times during the governance period – this account became ineligible. It is worth noting that the wallet balance was brought back above the original commitment but this doesn’t undo the decommit, and so the wallet remains ineligible until the commitment window of the next Governance period, when it can be used to commit to governance for that next period.  The vote cast by this wallet was thus negated, as this occurred during the cool-off period and did not count toward the final vote tally. In addition, this wallet is no longer eligible to claim Governance rewards.

There has also been a small number of other mid band ( > 1M Algo ) wallets that either did not vote or rendered their vote ineligible by allowing their balance to slip below commitment during the cool-down period. As with the larger single wallet event, these governors’ votes, if cast, do not count and none of these accounts will be able to claim Governance rewards at the end of the voting period.

If we now look at the eligible votes and how they voted, we see the following. The first charts show the final snapshot of the eligible voters, in terms of number of governors by stake size and stake by stake size.

As you can see from the charts, the split of votes between A and B were different in different categories. It is clear from the charts that all groups and sizes of commit voted for both options and no grouping ( neither so-called “whales” or “small holders”, voted exclusively for one option or another ). If we look at small holders and the largest whales, both groups voted A in clear majorities and in fact, Option A was carried in terms of voted Algo commit and total voters.

If we look at the vote over time, we see larger holders voting later in the cycle, and that on the 10th and 11th of November, the total commitment count switched from being in favour of Option B to Option A.

However, while the swap-over in committed Algo vote occurred during this period, from a simple count of governors perspective, Option A was never behind in the vote, although the vote gap narrowed over time.

For the community, it may be interesting to analyse further the period from 9th November to 13th November to see if there was an impact on voting, and potential vote preference change, as a result of governors choosing to publically articulate their view in support of a specific option. The most significant example of this was Arrington Capitol’s contribution and analysis on November 11th here.

In Summary:

The first voting period has some key takeaways for the Algorand community.

1. Decentralization

The scale and commitment of the vote by the Algorand governors is a clear indication that the Algorand ecosystem is on route toward complete decentralization. Governance at this scale has never been seen before within a Layer-1 blockchain and the momentum established by the first governance period is likely to bring yet more governors into the Algorand ecosystem, expanding the size of the electorate in future community votes. This broadening of the electorate over time can only further diversify the community and its inherent decentralization.

2. Vote Incentivization

Incentivizing the vote has been a hugely important pillar of Algorand Community Governance and it has worked. With the caveat of the test accounts that were in Governance period 1, we have seen a very low fall-out in committed stake due to failure to vote in this period. We have however seen fall-out due to accounts moving below the commit threshold during the governance period. We expect that many of these, including the example shown in the detailed analysis, were due to simple errors. For future voting periods, with fewer test accounts and greater familiarity with the commitment threshold requirement, we expect to see an even higher governor “turnout”.

3.Engagement

The first voting period has seen a wonderfully vibrant and engaged discussion in online forums ( Reddit, Discord, Telegram ) and social media ( twitter primarily ) around the relative merits of Option A or Option B. The relative merits of both options, the impact on the long term structure of the Governance program and on the ecosystem itself, were discussed in great detail, and with strong passion, for the various positions held. In a wider sense, the first vote has already led to a significantly more engaged community that now has the opportunity to not only display their commitment to the ecosystem through holding Algo but through their personal commitment to engage in key decisions of the ecosystem.

4. The Largest On-Chain Vote in Blockchain History

Finally, we should end this analysis by revisiting what must be described as a historic event in Blockchain history. Over 52,000 individual Algorand community governor accounts, committing over 1.8Bn Algo ( greater than $3Bn USD at the time of the vote ), took part in the largest decentralized, on-chain, governance vote seen in the blockchain space. And this is just the beginning !

Backup: