Piers R is an Ethereum early adopter, blockchain enthusiast and co-founder of stealth startup BXD.io.
In this post, he responds to last week’s opinion post by Washington, DC-based lawyer Stephen D Palley, setting forth a series of legal questions he believes innovators seeking to launch blockchain-based autonomous organizations should consider.
An Ethereum-based distributed autonomous organization (DAO) is a blockchain entity that operates according to a set of pre-defined rules that the members of the DAO help maintain and that they leverage to make collective decisions.
A DAO (which differs from other concepts like a DO) is principally created as a vehicle to achieve or maintain a shared purpose, and that will both receive and distribute funds (often in the form of cryptocurrencies or other blockchain-based tokens of value) and control actions designed to help promote or further the purpose of the DAO.
On the more mundane side, with no legal identity, owning physical assets, signing contracts with non-blockchain companies (real-world suppliers, contractors, etc) and tax issues can be difficult.
For example, it is not known:
Where an action is incidental to the purpose of the DAO, it is reasonable to expect a service provider (SP) will bridge this real-world gap (eg providing a one-time hardware design service).
But if something is central to the purpose of the DAO (eg ownership of the IP once the design work has been paid for), you are really going to want the DAO to be able to sign more than just smart contracts, especially if the DAO is to be able to realistically split from a SP (ie you don’t want the SP keeping an asset belonging to the DAO after the split).
On the more worrying side, with no independent legal identity for the DAO, should loss (financial or otherwise) be caused by a decision or action made by the DAO, the best choice in law is to sue all stakeholders, holding them collectively and individually liable for the decision they were party in making.
In UK and other common law jurisdictions, this is often called “joint and several liability” and it means that I can choose to sue everyone, but even if I can only find one person actually worth suing (ie they have the means/assets to pay), I can sue them for the entire amount I am seeking.
Then, it is up to them to find the other parties and make them pay their portion.
One example is if the DAO makes a decision or action that causes one or a minority of the members to experience a loss.
A simple example of this is peer-to-peer insurance: a DAO is created as a collective insurance pool.
After taking insurance premiums for a while, the DAO grows and hires a SP to help investigate some claims. A member of the DAO tries to make a claim, but it is rejected by the DAO members. The claimant believes the rejection to be groundless. The DAO also elects not to pay the SP as the members do not believe the service providers work is up to scratch.
Who does the claimant sue? Who does the SP sue?
While the terms and rules of the DAO may be incredibly well written, if they do not comply with changing local laws (eg consumer protection legislation or sale of goods and services rules), then it won’t matter how well the DAO is defined, or what kind of waivers you make people sign, it can still be sued.
Limited liability companies (LLCs) were created to help solve some of these issues. It gives a collection of individuals a legal identity that can enter into contracts and be held liable for its actions, but that limits the liability of those individuals to no more than the money they had already committed to the company.
In the UK, there are several forms of limited liability organizations – this post will focus on:
Generally speaking, a company limited by shares is designed explicitly for the making of profit for its shareholders. This is the most common form of company, and the total amount of money its shareholders can be pursued for (after net assets are taken account of) is the capital invested by the shareholders.
For strictly for-profit DAOs, the structure may work as follows.
The members of the DAO are shareholders and they must purchase shares in the DAO when these shares are offered for sale or when the company is created (as part of the founding share capital).
If further members wish to join the DAO, they must purchase shares in the company at the market rate, either from existing shareholders, or as part of a further capital raise. As the value of the DAO increases (providing dilution is less than value creation), so does the value of the shares in the entity.
Directors are appointed to the board by the shareholders – generally speaking those with the most shares are those that get a seat/choose someone to represent them on the board.
The main drawbacks with a company limited by shares for a DAO is that the corporate control of the DAO may be purchased simply by purchasing enough shares to have a simple majority.
This potentially puts the control of DAO-owned, non-blockchain based assets (such as IP, property, listed stocks, etc) with the majority owners of the company, despite the best efforts to avoid centralization of control in the DAO rules.
Companies limited by guarantee (LBG) are not designed explicitly for profit, and are normally formed as social enterprises or community organisations.
Instead of investing capital into the company to become a shareholder and thus looking for a return on that investment, companies limited by guarantee (LBG) are instead limited by the guarantee of its members.
This is so that, in the event of the company being closed down, the members are liable up to the amount of the guarantee only. Generally, this is a nominal amount like £1 per member.
For entities such as a DAO, LBG companies look very appealing.
For one, they are not explicitly formed for profit, and are instead formed to carry out an “objective”. No one person can control a majority of corporate decision making simply by buying it, instead you can make each new member of the DAO a member of the LBG entity as well, giving all a voice in both.
It gives a way of separating the issue of corporate legal control from the ownership/rights to profits in the DAO.
For example, a DAO could be incorporated as a LBG, but then issue and sell tokens to members that give them rights to profits made by the DAO (eg crypto shares such as those that startup Slock.it plans to issue).
This effectively creates a for-profit DAO, but it is harder to undo by a majority ownership-style attack.
For profit-making DAOs, it makes sense for the size of your membership or vote in the LBG and the size of your stake in the DAO to be linked for initial investors. Once that stake is sold on, however, the subsequent transaction does not add any value to the DAO.
Certain DAOs may then choose for membership rights to have an earn-back period to help reduce the incentive for speculative and malicious third parties to purchase shares for control of the DAO, as well as protecting the purpose and longer term members of the DAO.
To adopt the LBG structure and give legal personhood to a DAO, the following additions or changes to the standard LBG Articles of Association (AoA) would be good to consider.
All guarantee funds could be taken upfront on the joining of the member and held in a separate blockchain-based wallets that can only be transferred to a bank account owned by the LBG company in the event of a winding up or liquidation event. This will be a £1 equivalent at the time of the membership and could be refunded one year after the member ceases to be a member (eg after leaving the DAO).
This makes sure that, should the LBG need to be liquidated, it would not be necessary to chase all the members for their guarantee payment (although if the value of the collected funds dropped below the value of total members multiplied by £1, it may still be necessary to chase people).
Membership and voting in the LBG should be linked directly to membership of the DAO, and potentially, should also include a minimum DAO membership period (to avoid spamming the DAO with bogus members to get control of the LBG entity).
This means that to become a member of the LBG, you must have been a owner or member of the DAO for one year period before membership is approved. No director may directly appointed another director. Vacant spaces for directors are either proposed by the board or by at least 5% of the members.
This proposed new director is then put to a confirmation vote by the members of the DAO. Any positive voting is problematic at scale (ie 5% of 1 million members is still 50,000 votes). Instead, I propose that instead a negative vote is put in place: the proposed new director is automatically approved after a minimum review period of 10 working days unless 5% or more of members vote against the approval.
The number of directors should be capped at a maximum (suggested nine) to avoid too many people having authority to act on behalf of the LBG. This uneven number is suggested to avoid director deadlock in decisions that are split.
Directors should also be restricted in their ability to bind the company individually, with any decision requiring the movement, divestment or purchase of assets equal to 10% or more of DAO+LBG assets to require a vote of the board and signature by two directors.
Directors should be selected on the basis of relevant skills and experience and be rotated on a staggered four-year maximum term to avoid the entire board from being replaced in one go.
While the above seems complex and bureaucratic, it is important to note that most of the liquid assets and work conducted by the DAO+LBG combo will be held and controlled directly by the DAO (ie in smart contracts and cryptocurrency).
The DAO will not need to follow anything other than it’s internal logic and rules to deal with those assets. It is only when the DAO needs to contract, purchase or sell in the real world, in a scenario in which a SP cannot and should not act, that the LBG structure and it’s directors would need to act on behalf of the DAO.
Network visualization via Shutterstock