Burning Algos Proposal for Period 8

This is a follow up to this post.

Proposal

Burn 100M Algos monthly (up to 2B) as long as Algorand is not one of the top ten coins. (The definition of top 10 coins can be adjusted as necessary)

Rationale

The current position of Algorand in terms of market cap severely hurts its potential growth (relative to what it could/should be). Investors, builders and the public are disproportionately attracted to top coins as they are perceived as more stable and future proof than smaller coins with larger communities. Becoming one of the top ten coins wouldn’t happen without a (significant) price increase. Given the current constant sell pressure, the number of Algos come to the market and not that promising future prospect about any major change in this dynamic, it is less likely that Algorand outperforms (if not underperforms as history suggests) the market.

Q&A

1. I don’t agree with the problem statement. Algorand is growing at a reasonable pace right now (happy days!) and in the next few years the situation will improve.

A: This argument is a risky one as it doesn’t compare the current state to what Algorand growth could be. Moreover there is a good chance that opportunities that exist today may not exist in a few years. One good example is that other chains can catch up with the Algorand technical advantages given more resources they naturally receive. Ethereum migration to PoS is a good example.If anything this proposal shouldn’t harm (or list them if you disagree after reading the following questions) but accelerates the growth by winning the attention of millions of people, which brings much more resources to the Algorand ecosystem.

2. I agree with the problem statement but don’t like this particular proposal?

A: Please provide a different proposal with the rationale behind it that addresses the same problem.

3. Doesn’t this proposal impact the foundation’s Algo holdings?

A: The proposal reduces the foundation’s Algo holdings in terms of number of Algos and not necessarily the dollar amount as the price increases, which is what mostly matters for the foundation operations. The ultimate job of foundation is growing Algorand ecosystem and community. Moreover how the holdings are used is expected to be decided by governors.The network fees, though small, are sent to a foundation account. Eventually the fees should be sufficient for the foundation and network costs.

4. Doesn’t this proposal impact the governance rewards?

A: It might impact it. There are already plans to reduce the rewards. For most people rewards are a compensation for the Algorand inflation and if there is less (and eventually no) inflation even with lower rewards they probably continue to participate.

5. Isn’t burning Algos a bad practice?

A: Burning Algos has happened in the past. Moreover, 10B is just a number and if really necessary in the future new Algos can be minted with the governors’ vote. The same way that the transaction fees might need to be adjusted.

Reddit

2 Likes

@Massimo @Adri for a better visibility.

A burn seems like a waste to me. Wouldn’t it be much better to allocate more funding instead to projects and programs that truly stand a chance at improving the overall attractiveness of Algorand and let demand do the rest?

7 Likes

Why waste? Isn’t the dollar amount that matters (for the majority of cases)? 10 Algo @ .1$ has the same value as 1 Algo @ 1$.

1 Like

This is true, but burning $ALGO irrespective of price is its own form of dilution I think, maybe I’m crazy. I would rather have that available in the future for when the price is higher and thus fund more projects with less. To me, it seems like this would have a ripple effect vs just a reduction in total supplies’ one-time effect.

1 Like

Algos will be available as less are sold because of the higher price.

No action would be perfect but the question is whether the problem is important and urgent enough to be addressed or not? My answer is yes. The above is one proposal. Other proposals are welcome.

agree with @averagezen burn is just a waste. it has become a common thing to make it seem like a token should be worth more after it but imo its just marketing and doesnt achieve anything really

5 Likes

I would agree with this - this was my first thought when I read the proposal.

1 Like

It changes the future prospects in particular the sell pressure as the Algo entities have to sell less number of Algos for their operations. It also means there is not that much remaining Algos for accumulation and if you want to take a position cannot delay it for a few years.

I don’t quite understand the waste argument. There is no shortage of Algo tokens. At the end what matters for Algo entities is the dollar amount not the number of Algos they hold.

1 Like

you seem to think that if we burn x% of supply then the price will rise accordingly. this could happen in the short term (some kind of hype) but i dont think it will help in the mid to long term. imagine what could be build with 100M ALGOs, if we dont use this to attract new and exciting projects to come to algorand or help the devs already on here then its a waste imo

6 Likes

No, that’s not the argument.
The argument is that it changes the future prospects of Algorand inflation and new Algos that hit the market (currently ~400M + inc sales per year). This creates a better balance between bids and asks and allows the price action to recover a bit or at least doesn’t constantly go lower. Just this, the fact that price is not constantly going lower attracts more buyers. Note that this is something that is being collectively understood by different people so everyone knows that there is less available supply at current prices and if they want to build a position they may not have a long time. All of this should create a positive feedback loop compared to the negative one we are in and there is no foreseable future that is changed.

1 Like

While I understand the concern regarding current market position and the attractiveness of being among the top ten coins (ranked by Market Cap, which is widely used but still a questionable metric), I believe that focusing on token burning may not provide the sustainable solution we’re looking for. In fact, it could be seen as an artificial and desperate way to stimulate short-term price action, which might not necessarily translate into long-term growth or stability.

The main issue I see with token burning is that it doesn’t fundamentally change anything about the underlying technology or the utility of the Algorand network. It’s more of a financial maneuver than an actual improvement or innovation.

Burning Algos would be akin to burning AWS credits (if they were traded on the market), which wouldn’t necessarily improve AWS’s service or reputation, but could impact its operational efficiency.

Furthermore, token burning could potentially create bad press and give off an image of desperation, which could undermine Algorands credibility and trustworthiness in the long run. Why would a chain that works need to resort to market gimmicks such as burning tokens? We should aim to attract users and investors with the merits of the technology and the utility of the network, not just the price of the token.

I would argue that a low Algorand price could even be beneficial for builders and innovators, as it lowers the cost of using the network and incentivizes experimentation and development. This could lead to more adoption and usage over time. Unfortunately the current markets tend to favor hype over usability, but being right doesn’t always mean being the majority.

Algos would be better spent on grants to support developers and projects that contribute to the Algorand ecosystem. This could encourage more innovation and growth, moving more supply into the hands of the people directly responsible for price action.

I think it’s better to bet on the tech than to focus on short-term price action. If you are here purely for the price action, you might be in the wrong place.

8 Likes

I agree with pretty much everything you said here. But I would add that burning Algos is more akin to a share buyback, which is pretty much a dividend without tax liability. We area already giving out a dividend via Governance, so burning is just a way to do largely the same thing with more favorable tax incentives and without requiring users to participate in Governance. There are a number of problems with this strategy.

  1. As you said it does nothing to improve the fundamentals of the tech or ecosystem.
  2. The protocol isn’t profitable. What sense does it make to hand out a payout to existing holders? Like any early stage technology investment, its going to take time before you see any returns and even then dividends are way out. Google and Amazon still don’t offer them and Microsoft took over 15 years to start offering one.
  3. Even with that, we are already handing out free ALGOs via governance and there has been a ton of discussion about why that has been somewhat negative for the ecosystem at large. Increasing how much free money is handed out (via burning or governance) mostly incentivizes ALGO to get concentrated into exchanges who take a cut from their users.
  4. I disagree with the idea that attracting more investors will attract the kind of development we really want. Lots of retail interest will attract developers who will pump out more scams, rushed (insecure) projects, and generally shit that we see on the other chains. It won’t attractive the adoption that really long term matters which is going to come from existing major institutions (Banks, regulators, etc).

Algos would be better spent on grants to support developers and projects that contribute to the Algorand ecosystem. This could encourage more innovation and growth, moving more supply into the hands of the people directly responsible for price action.

I think it’s better to bet on the tech than to focus on short-term price action. If you are here purely for the price action, you might be in the wrong place.

:clap::clap::clap::clap::clap::clap::clap::clap::clap::clap::clap::clap::clap::clap::clap:

3 Likes

@grzracz

I don’t disagree with some of the listed cons. However, the question is whether the trajectory of the current trend is promising in the competitive landscape. Both entities need funds that are directly coming from the community. DeFi needs liquidity that comes from the community. Any aspect of blockchain growth is dependent on the people who use it, whether as an individual or entity. It is a strong community that keeps the blockchain alive in a hostile environment and bear market. The continuing declining Algo position negatively impacts the community growth and confidence. It has also created a bad reputation that prevents new people from joining. At the same time other and new blockchain are emerging with an enhanced tech, new ideas and thoughtful strategy.

What I’m seeing in the market is not good. The market has lost its confidence in Algorand and that means new capital, which is the oxygen for growth, is not allocated.

The key question is what alternatives you have in mind that could reverse the current trend? I’m willing to hear other ideas.

1 Like

I disagree with the proposal. We shouldn’t be changing the supply on a whim based on an arbitrary position of a speculative leaderboard. Let’s keep it as it is defined today and find a way to enter the top 10 by walking a different path.

5 Likes

Provide your concrete strategy please? Also see the above comment just posted.

We all should give it more thought and come up with less intrusive strategy. Reacting and doing arbitrary changes to the emission because of the current position based on marketcap is just not a good thing for the future confidence in the asset.

1 Like

Sure. But that’s not a concrete strategy. What is the process and timeline? also who drives it?
You know well that just saying that “we all should do …” probably means nothing will be done.

I think doing nothing is still better than what you suggested. Again, arbitrary changes to system variables because of the current price are a terrible idea.

2 Likes

Still no strategy, only talks (no insult, just to be frank).
BTW, if you call it arbitrary (not the exact numbers but the approach) how do you describe many past distribution decisions like accelerated vesting?