Life, the Universe, and Decentralization
In a newsletter last week, Mike Ippolito suggested that the execution of decentralized governance in crypto often follows a one-size-fits-all approach.
Many projects and DAOs decentralize everything, as quickly as possible…often to satisfy a particular idealistic attachment to the concept, but sometimes simply to avoid the Wrath of Gensler.*
Whether this is necessary (or wise) is up for debate.
Mike argues that progressive decentralization over time may be a more nuanced approach; and that in fact, perhaps we only really need it at the protocol level anyway. Why would a game development studio, for example, need to be decentralized?
My interest in this space has always been as a pragmatic decentralist. I'm an ardent capitalist, but I believe that centralizing too much power in the hands of too few wealthy people is A Bad Thing.
I'm also committed to the notion of the social contract, but I believe that centralizing too much power in government is likewise A Bad Thing.
Which is why I postulate that the argument for decentralization is in fact an argument for decentralization where it matters (hence the pragmatism).
Some institutional structures are optimal when centralized, and that's just okie dokie artichokie.
As Mike suggested, a startup is likely one of those structures. Early-stage organizations might have a couple of highly-dedicated founders who share a common vision. It's often their passion and sense of mission that drives the nascent organization forward. Their incentives are, ideally, closely aligned. They develop further incentives to bring employees onboard.
But what if that early-stage company is decentralized from the outset?
Now you need to find a whole raft of people who share that vision and mission. (And if you think finding one co-founder is hard…)
If there is misalignment in incentives — because Alice needs to finance a mortgage, Bob wants to prove his consensus mechanism works, and Cathy is dedicated to social good — then decision-making becomes fractured, and the organization cannot move forward with the same focus.
In which case, the law of the corporate jungle dictates, it will be out-competed by a centralized competitor.
In this example, decentralizing a startup is not only counterproductive, it's also comparatively unimportant — at least to those outside it. This fledgling organization has just a few people. It's not world-changing. It's not even market-changing yet.
But imagine that organization is Facebook, some 20 years later.
Mark Zuckerberg essentially wields total control over Meta's direction — thanks to his Class B stock and, with it, outsized voting rights — despite its progression from private to public company (which, in most offerings, leads to a degree of decentralization, where the opinion of the shareholders actually matters).
Is this healthy? The pragmatic decentralist in me says no.
It's not healthy for Meta (because Zuckerberg can opt to pursue business lines that may not make sense), or for the shareholders (whose recourse is essentially limited to selling their stock), or for humanity (because one man controls a platform that affects the lives of billions of people, and he doesn't seem particularly inclined to use that power for good).
So somewhere between startup and monopoly, there's a case to be made for decentralization.
At some point in this journey, it began to matter that Mark Zuckerberg is unaccountable.
And although Meta's example may be extreme, that demarcation is what I believe we should be exploring in governance right now.
When does decentralization matter?
We know that it's hard to move Ethereum governance, because it's pretty decentralized and the various constituents and stakeholders are rarely moved to action by a common aligned incentive. (Or at least, they rarely agree on how to execute changes that result in one.)
We also know that thanks to VC token locks and investments, some ostensibly-decentralized organizations, particularly in DeFi, are actually beholden to majority token owners who are essentially democratic dictators.
So perhaps the ultimate question for those designing governance mechanisms is how to implement pragmatic decentralization that is appropriate both to the maturity of the organization, and also to its 'criticality' — the importance of that organization to those affected by it.
What does a decentralization spectrum look like?
The answer is undoubtedly 42.
But I have a nagging feeling that the question could do with some refining.
*Avoiding the Wrath of Gensler may not be possible, depending on the price of milk, the position of Venus in the night sky, and which side of the bed he woke up on this morning. This concludes today's interpretation of the Howey test.
No comments:
Post a Comment