I came across an interesting paper on the website of the Harvard Kennedy School’s Belfer Center for Science and International Affairs that I think is worth picking apart in some detail. It bears the title “Toward Equitable Ownership and Governance in the Digital Public Sphere,” and it is a wonderfully clear example of that old idiom: “if all you have is a hammer, everything looks like a nail.” In this particular case we have a group of academics bearing a hammer labeled “DAO tooling” coming after co-ops with an unsettling gleam in their eyes.
DAO is an acronym meaning Distributed Autonomous Organization, and it comes to us from the cursed world of cryptocurrency (if that fact is new to you, you might want to stop reading this right now and go read this first, lest what follows be utterly incomprehensible). The authors admit that there is no agreed upon definition of a DAO, but for our purposes that doesn’t really matter. What matters is their suggestions for how co-ops can use “DAO tooling” to increase their growth and competitiveness. These suggestions are, without exception, terrible. The reason for that is not necessarily the suggestions themselves, but the repeated emphasis on using cryptocurrency smart-contracts to implement them. To give you a taste, this is from the Executive Summary:
This paper explores how newly developed DAO tooling could help co-ops compete in the online economy. Specifically, we outline how DAO tooling could provide co-ops with:
Effective Voting—DAOs test novel and varied governance systems that appear to offer some unique benefits not available to legacy organizations. These include systems such as reputation-weighted voting, holographic consensus, conviction voting, and quadratic voting.
Increased Member Engagement—DAO tooling allows organizations to gamify milestones for their members, autonomously tracking the progress of members toward their goals, and rewarding them based on behavior. This encourages members to stay on task both for short-term gain and to develop a strong reputation in the long term.
Here we have two common issues that cooperators do spend a good bit of time discussing, debating, and experimenting on. The proposed solutions, however, are way off the mark. For instance, while it may actually be the case that any of the alternative voting schemes mentioned may be better suited to purpose than whatever decision-making proceedure a co-op already has in place, there is no need to make use of a blockchain to accomplish any of them. While it may be possible to implement these voting schemes on the Ethereum blockchain, that doesn’t mean the Etherereum blockchain is necessary, or even useful, in doing so. And here I'll mention a fundamental aspect of DAOs that the authors don't discuss, and that rarely gets mentioned by crypto advocates: the “gas fees” that must be paid any time the DAO interacts with the blockchain ledger. Here’s a description of how they work (because it’s crypto nonsense, you know it’s going to be needlessly complicated):
Gas price is the amount of ether (ETH) that a user is willing to pay for each unit of gas, measured in gwei (1 gwei = 0.000000001 ETH). The higher the gas price, the faster the transaction or smart contract will be confirmed, but also the more expensive it will be. The lower the gas price, the slower the confirmation, but also the cheaper it will be. Choosing the right gas price depends on the urgency and importance of the transaction or smart contract. A DAO can use tools such as gas trackers, gas estimators, and gas limit calculators to find the optimal gas price for its smart contract, based on the current network conditions and the expected execution time. [emphasis added]
Imagine running a governance system where your co-op was charged money every time someone cast a vote. Why would anyone ever do that? They wouldn’t, of course, which is why I’m not terribly worried that this will catch on with co-ops anytime soon. Because while crypto advocates try to obscure the price of fees by quoting everything in their made up “currencies,” a quick conversion shows the network’s cut to be quite high in dollar terms, not to mention their abysmal wait times.
So making a change to the ETH blockchain (which, remember, is how DAOs function) currently costs a minimum of $2.08, and you’ll have to wait around 3 minutes for the changes to take place (the price went up to over $3 while I was writing this). Meanwhile, I can make a change in a shared Google Docs spreadsheet and it doesn’t cost me anything at all and happens instantly, and Google keeps a record of the change that everyone can see.1
Next, the authors tackle the eternal bugbear of member engagement. Here they offer DAOs as a way to “gamify” participation which is another thing that can be done perfectly well without involving the blockchain and its associated gas fees. Rather tellingly, in the full paper the only example they give of this is Mochi, a Slack and Discord bot combined with a smart contract that automatically takes money out of your Ether wallet if you or your friends fail to post a daily affirmation. It is incredibly dumb and seems like a really good way to ruin friendships. Fortunately, the website currently only lists five users, which is actually rather high, if you think about it.
The next problem the authors claim to have a “DAO tooling” solution for is “Predictable Compensation/Patronage.” The problem here is that I don’t think this is an actual problem. The authors themselves provide no evidence that it is, simply stating that have problems determining pay rates with no reference or example provided.
Co-ops often have issues determining how much each member should be compensated for their work. Distributions to members of a co-op are proportional to their labor, efforts, and success in using the co-op, also known as “patronage.” The more a member patronizes a co-op, the more services that member receives, and/or the more of a co-op’s earnings are allocated to that member (Walden and Spelliscy, 2020).
That citation at the end, that one might assume would support their contention of this being a real problem, instead leads to a 2020 post about how DAO founders can use co-op structures to avoid securities regulations (which is, I believe, the main reason DAO people are taking an interest in co-ops in the first place). Their example of "DAO" tooling in this sphere is something called "Coordinape" which is just a system for showing appreciation among co-workers. As the authors describe it:
Coordinape...allows members of an organization to allocate tokens, called GIVE, to other members of the organization in recognition of value creation, providing insight into compensation decisions.
Assuming that some type of tokenized recognition system would be of benefit to a co-op or co-op network, why involve a blockchain? Like most of the solutions the authors propose, this one would be better accomplished in a way that doesn't involve paying gas fees and meddling with tokens that may or may not be considered unregulated securities.
Next we come to “Transparency and Accountability” where once again we have the recognition of a legitimate issue combined with a frankly bizzare suggestion for a solution.
...on-chain DAO activity like token voting, treasury management, and the payment of salaries and subsidies for public goods can be recorded on an immutable public ledger viewable by anyone with an internet connection.
I know of very few businesses – none, if fact – that would be amenable to having their books viewable by literally anyone with internet access. One has to wonder exactly which co-ops the authors consulted in conceiving of this “fix.” And once again, there is already a well-understood and widely utilized solution for creating transparency in managing budgets, salaries, and the like: it’s called “open book accounting.” And remember, since all “on-chain DAO activity” comes with gas fees, this suggestion amounts to, “hey, why don’t you pay to make sure everyone, not just your members, can dig through your books.” It’s bizarre, frankly, and points once more to the whole issue of hammers and nails that runs throughout this paper.
The next non-nail that the authors attack with their DAO hammer is “member accountability.” They write:
A user’s reputation can automatically change based on the quality and quantity of contributions to a community. Reputation can be non-transferable, tied to a particular individual or organization, and recorded on an immutable public record. These measures can make accountability in a DAO easier to maintain than in a traditional co-op, where it may be hard to identify rule-breakers and hard to enforce accountability as the organization grows.
Again, here the authors have identified a relevant issue and offered a very bad solution for it. All organizations have to deal with accountability, and it is a surety that accountability mechanisms and systems could be improved in some co-ops; but adopting a system that makes accountability issues – which are personnel issues – public to everyone on the internet is not only a very cold and mechanical way of addressing the issue, but it might also have a hard time getting by the legal department, for the same reasons that most companies keep personnel records confidential. And that word “immutable” is especially problematic in this context, as being immune to change here simply means everything everyone does will go on their permanent record (which will be publicly visible to literally everyone). I could riff here on god complexes and the dystopian uses to which such systems could be put, but fortunately the suggestions are so impractical that I don’t think I need to bother.
The final issue the authors address is another perennial one: how to provide more outside financing for co-op development.
Co-ops are at a significant disadvantage to corporations when raising the funds necessary to scale and compete. They can generally only raise capital by direct contributions through membership fees, by agreement with members to withhold a portion of net income based on patronage, or through retention of a portion of sales proceeds for each unit of product marketed (Walden and Spelliscy, 2020).
The authors are either unaware of existing co-op funding mechanisms like CDFIs, regional loan funds, and preferred shares (which seems unlikely) or they have made a conscious decision to not mention the existence of a large part of the co-op funding landscape in their paper. Why would they make such a decision? You would have to ask the authors to be sure, but I would guess it might be because including them highlights how unnecessary “DAO tooling” is in this area. Which is not to say that we couldn’t use more funding, and more varieties of funding, for co-op formation – we could – but what the authors are selling ain’t it. And they admit as much, themselves, before suggesting some more bad ideas from the DAO toolbox.
While securities laws and lack of legal clarity make it difficult or impossible to reliably use blockchain tools for fundraising, particularly in the United States, DAOs have started experimenting with alternative forms of funding which could ultimately be relevant to co-ops. These include selling or airdropping tokens to bootstrap treasuries.
If someone can explain to me the difference between "using blockchain tools for fundraising" and "selling tokens" I would love to hear it. They are, of course, one and the same thing, and as they mention, securities laws don't necessarily look too kindly on the practice. Just consider: their big idea for helping co-ops raise money is for those co-ops to sell unregulated securities. The mind boggles.
This is where I’ll remind you that this paper bears the imprimatur of Harvard University. If it didn’t I probably wouldn’t bother dissecting it as I’ve done here. It would just be another paper by another written by a group of academics with Silicon Valley stars in their eyes. The web is full of them. But coming with the imprint of an “elite” university lends a certain credibility, and may therefore make these suggestions seem plausible to those without enough prior knowledge to see through the smoke. For that reason, I’ve taken the time to go through just some of the howlers in this paper and explain why that’s exactly what they are. I hope at some point this fad passes entirely, and its promoters move on to other obsessions, but until then I’ll have to keep writing these occasional missives, as I don’t see many other people in the co-op movement saying anything critical about this stuff.
And we need to be critical of it. If we are not, not only will poorly thought out solutions to co-op problems get given a credence they in no way deserve, but those solutions that are already here in existence – getting along quite well without any need at all for DAO tooling or blockchain anything – will end up being erased, or at least downplayed.
The paper provides a perfect example of this in claiming that cooperatively owned Mastodon instances would make a good fit for their DAO solutionism. They mention Social.coop, which is a cooperatively-owned Mastodon instance with fully transparent accounting and decision making, that has had few problems in scaling immensely since it’s inception. I’m a member of the co-op, as is one of the authors, as is noted in the paper. Tellingly, though the authors choose to focus on blockchain solutions to problems that might occur in the future, while ignoring the blockchain-free solutions that we have actually developed to meet the problems that we’ve actually faced (which are surprisingly few, given what we’re doing). We make use of Loomio and Open Collective for decision making and accounting, already accomplishing much of what “DAO tooling” promises to deliver but at a much lower price.
And finding Social.coop mentioned here was even more irritating to me, as the overwhelming sentiment from members on both our Mastodon and Loomio discussion threads has been pretty uniformly anti-crypto. I can count on one hand the number of members I would expect to be excited by adding “DAO tooling” to our already perfectly well-functioning co-op, so to find ourselves being put forward as a candidate for this crypto-fication feels to me like a very real invisibilization of who we actually are, in the service of selling something that many of us despise2 .
If you’ve made it this far, congratulations and thank you! And if someone ever tries to sell you on joining a DAO, my suggestion is that you just back slowly away and go find yourself a nice co-op...and then send them this link.
- 1Advocates might argue that you can batch transactions to save on gas fees, but that just means using some "off-chain" mechanism to do the actual recording, which defeats the entire purpose of using a DAO in the first place.
- 2I realize that I am relying on anecdata for this assessment, and there may be more crypto enthusiasts among the membership than I am aware of...but if there are they sure keep it to themselves, which is not something crypto enthusiasts are generally noted for.