In this post I’d like to cover some of the things that we're thinking about at Bunches as we're designing a fungible token for onchain consumer utility.
To be clear, we are not designing a token as a financial asset for onchain consumers, and so a lot of the thoughts that are here shouldn't be applied if that's your use case.
Our use cases involve sports fandom and reputation, both of which are what I'd refer to as "soft assets" (as opposed to hard financial assets). That said, they're still quantifiable assets that can accrue value from multiple parties who can be aligned with a single onchain utility token.
Utility tokens in this way don't necessarily represent financial assets to be transacted, but they are economic assets that quantify non-monetary value within an ecosystem.
For the record, just because our use case is sports fandom & reputation, I hope this post could be helpful for a range of soft assets across verticals: reputation as a contributor, cultural alignment, shared ownership in a non-treasury good, governance power, social capital accrued via follower graph strength, etc.
The considerations here really only apply to fungible tokens (ERC-20), but there may be lessons that can be abstracted to non-fungible tokens (ERC-721) or hybrids (ERC-404, ERC-1155) as well.
Disclaimer: this post is not intended to be a comprehensive look at designing utility tokens in general but rather is a look into how we’re designing a utility token at Bunches intended to quantify sports fandom. I truly hope it’s helpful to others thinking through similar use cases.
Supply Size
Let’s open with a broad blanket statement that is almost categorically false, shall we?
Increasingly, I’m convinced that consumer utility tokens should be infinite in supply.
For technical purposes, unlimited or infinite means 2²⁵⁶ - 1, as that's the largest value that totalSupply, which is an unsigned 256-bit integer, can be set to in a contract that meets the ERC-20 standard.
Unlike a financial system where there is a fixed money supply (M0, M1, M2, etc) controlled via a monetary policy by the issuer (the US Federal government, BRICS, the Bitcoin standard, etc.), consumer systems are comprised of infinite, non-zero-sum value exchanges.
Put another way: there is no theoretical limit to social capital (n² relationships, where participants can be infinite). Or reputation scores. Or power as implemented in governance.
In cases where soft assets seem to be limited, it’s only a practical limit (for instance, a social network can only be as large as the number of humans on Earth) or an infinite limit confined to an arbitrary scale (Uber driver and passenger scores - surely not all 5-star drivers are truly equivalent!).
I’m of the belief that the quantifying mechanism (tokens in this case) should mirror the value which is being quantified: financial or otherwise.
This is the point where farmers and flippers are both staunchly in disagreement. They behave and believe that the only way to build value is through systematically-imposed scarcity; they often assert that a fixed, limited supply leads to an increase in financial value and therefore a better system.
And that’s ok. I just disagree.
But even if you are in agreement that a token quantifying soft assets should be unlimited, it doesn’t mean that there aren’t drawbacks.
We absolutely should be thoughtful around the economics, even for soft assets. We need to take into account the fact that there may in fact be financial value at some point in the future.
Just like in the real world, hard assets may be traded for soft assets.
Avoiding the devaluation of soft assets is just as important as it is for financial assets.
Put another way, inflation is generally bad.
Deflationary Utility
The biggest drawback of an unlimited utility token is not that it’s more difficult to financialize, but rather that such a large supply risks devaluing the thing you’re quantifying.
If there’s an unlimited amount of reputation, what does it truly mean to be in the top 10%? It could just mean that you’ve been around longer than the next person. Or that you’ve curated more cultural assets. Or that you’ve amassed more governing power simply by not exercising it.
So how do we handle inflation? By building deflationary properties into the token.
There are two main tokenomic mechanisms that we can use to control an unlimited supply token, one of which you may be familiar with and another that may be new.
TRANSACTIONS
The first mechanism that can be used to throttle supply is perhaps the most obvious: transactions themselves. This is also the most familiar, as it’s what Ethereum uses to control its own utility of compute power.
By building deflationary (“burn”) mechanisms into the usage of the token, you have some control over the supply as it’s used.
Whether it’s the actual spending of the token to perform an action or it’s a blanket expense to perform any transaction, building deflationary mechanisms into transactions are a clear way to control a token with infinite supply.
I won’t belabor this category, as you’re most likely already familiar, but here are a couple of examples:
Voting to kick a user “spends” your reputation
Proposing a change to a DAO uses your governance token
Burn fees for every transaction (similar to Ethereum)
DEMURRAGE FEES
I was recently listening to Stripe Press’ fantastic infrastructure podcast, Beneath the Surface, and the episode with Ryan Peterson (of Flexport fame) sent me down a shipping container rabbit hole.
Did you know that if a container sits for too long in port or on chassis without being moved, the owner of that container is charged a fee?
This fee is called a demurrage fee, and applies when the logistics network starts to choke due to the inactivity of a single actor. As it turns out, demurrage fees are also a concept in traditional, real world fiat currencies that require storage: precious metals, paper, etc.
That is, when resources are being accumulated and are underutilized, costs are incurred by those not using the resource.
This demurrage fee concept can also apply to social & utility tokens.
Like international goods, soft assets like culture, power, and reputation are meant to be in motion.
In a thriving economy, value is circulated not hoarded.
If someone “hoards” reputation or governance or culture, and it sits underutilized, those actions eventually choke the network, which is meant to be constantly active.
So build demurrage into the system. If someone sits for too long, they get charged fees. Interest, if you will. And those tokens slowly drip back into circulation.
One caveat here: you have to be aware of the incentives that demurrage introduces into the system. Will people contribute in low-quality ways in order to avoid demurrage? This kind of dynamic has to be mitigated
Initial Token Allocation
At Bunches, we've opted to build a product and network first. And an economy second.
Therefore, another category of questions we're working through internally is that of kickstarting our economy: what's the entry point for users?
There are two ways to think about this: start the user at zero, and they earn their way as an actor in the system or start users on an equal playing field, and they gain or lose status as an actor in the system.
We're leaning towards the latter, where users start with an initial reputation score/balance, and can then earn or lose based on in-app behavior. Why?
Frankly, because this is the behavior we want to see in the world regarding social capital.
I believe that if you go around not trusting anyone until it's earned, your human experience is going to be lacking.
Assume reputation until proven otherwise...but when proven otherwise, there should be a steep cost.
Defending Soft Assets
Sybil attack. Two words that strike fear into the heart of tokenomists all over the world.
The stakes are arguably higher with soft assets.
In a world where people aren’t seizing the network for financial gain, they want power or infamy or cultural control.
And often easier. There is no need for a 51% attack if you have a sufficiently large minority that you control on a cultural platform.
You could say that this is what’s happening on many social networks, with Meta and Twitter being overrun with pseudonymous spam accounts and fake news.
Simultaneously, eclipse and Finney attacks are much more impactful when the very thing being quantified and exchanged isn’t money but information. A “double spend” of information is trivially executed compared to a double spend of currency.
How then should we defend consumer networks?
You can’t.
~fin~
Just kidding.
The short answer is similarly to how you defend financial networks from similar attacks.
Ensure costs to attack are sufficiently high enough to deter malicious actors.
Run tests and model usage with malicious actors. Adjust costs accordingly.
Perform identity validation within reason (phone > email or username login as a start, for example).
Use onchain social graphs as a trust proxy.
Utilize proven Sybil-resistant methods at the point of use (SumUp for content voting, EigenTrust for peer-to-peer reputation, etc).
Each one of these in isolation isn’t enough. But together they provide layers of protection for your network and for consumers.
What's Next?
Well, we're obviously implementing a lot of these ideas and thoughts at Bunches. Maybe I'll report back with learnings or lessons. Or you'll just see it onchain. ;)
Want to learn more, or just chat about this? Ping me! I primarily hang out on Farcaster these days, so feel free to send me a reply (thinking in public is great!) or shoot me a direct cast there.
Otherwise, thanks for reading! 🙏
Feel free to collect, tip, or otherwise share this content with others you think would appreciate it.
- Loading comments...