Part II of DAOs: New Horizons and Challenges in Depth

Part I - Part III

Understanding “The DAO”

The DAO is a special case of a financial DAO that is designed to make investments in external entities. It came about when Slock.it, a company that is building a “Universal Sharing Platform” for Ethereum that will govern access to resources such as houses and bicycles using specialized electronic lock devices, decided to raise money from the decentralized community to fund R&D. They realized that using a decentralized crowdsale to fund a corporate entity resident in Europe that would develop hardware would be more challenging that funding development of, say, a purely decentralized system such as a prediction market. Since significant extra work would be involved they decided they would build a generic decentralized crowdfunding platform that might be used not only by them, but by any venture that wished to raise money from the crypto community — and this platform is now known as The DAO. Their plan was to launch The DAO, and then submit a proposal for their own funding (at the time of writing, The DAO has not yet begun considering proposals, but it is considered highly likely that Slock.it will submit a proposal and gain significant funding).

While The DAO’s launch plan was very professionally executed by the Slock.it team, nobody anticipated the degree of demand to make investments into its funding pool in exchange for voting tokens. At the time of writing over $130MM has been invested, much to the bemusement of some pundits.

The misunderstandings are illustrative since even sharp and adroit commentators as Emin Gün Sirer sometimes miss the point:

Regarding Emin’s comment, while it is true that the operators of platforms such as Kickstarter and Angels List haven’t absconded with funds — at least in the US — the primary reason for The DAO’s success was not that it solves this particular problem (I will note before leaving this point however that Bernie Madoff’s fund lost $18 billion before his fraud was detected, Mt.Gox decimated the fortunes of many crypto investors and operators of a P2P lending platform in China recently absconded with billions, reminding us that an unregulable decentralized financial system’s facility to store wealth in manner that makes grand malfeasance more difficult remains hugely significant). Nor was it, particularly, that trusted fund mangers might be disintermediated and their fees avoided.

The DAO has succeeded so spectacularly in its fundraising precisely because there are many thousands of people around the world who want to deploy their capital freely as a community without the yoke of geo-specific regulations hanging around their necks that would make this enterprise entirely impossible without decentralized technology. This community does not care if they are qualified investors and they do not want the “protections” provided by the existing system. For the first time in history, with relatively negligible cost, a worldwide community of individuals joined by similar outlook has been able to come together to in a global investment fund that manifests as a protocol that hopefully can stand above the vicissitudes and restrictions of different regulatory regimes — in a word, it offers the promise of freedom.

The DAO and The Challenges

Freedom is a heady thing that can make you drunk and make mistakes. The DAO does indeed represent a revolution, an airburst that announces a new phase in history in which assaults can be made on monopolies and areas otherwise guarded and protected by regulations in which people might organize themselves and their capital more independently of their masters. But this new world is relatively undeveloped and full of pitfalls. Firstly, in the case of The DAO, its backers may unwittingly have adopted a mission far more challenging than investing in the decentralized systems that are their bread and butter where returns and revenues might be derived from the operation of relatively predictable protocols, math and smart contract logic. Instead, they will be investing largely in off-chain entities; normal companies that will be expected through social contract to return a portion of dividends (profits) they eventually generate. Emin Gün Sirer’s other objection made the point:

Far from being transparent like smart contracts, off-chain entities such as corporations are highly opaque in nature. Once capital is invested, even if they become profitable directors can pay themselves large salaries or rack up wanton expenses to ensure profits are minimal. And, if for whatever reason they should decide not to honor their social contracts, or perhaps sell assets they developed using investments to friends for a song, social contracts will prove a poor tool with which to recover funds using the traditional legal system. But for now let us assume exemplary investments decisions are made — and we will get to critical issues with the communal decision making mechanisms in a moment — and that investees behave themselves. The principle of investing to gain a share of future profits rather than a legal share of ventures remains problematic.

A key problem is that it does not provide for investors to realize gains in the underlying value of their investments as they arise. To understand, consider investors in decentralized projects such as Ethereum that purchase its value token, such ether (ETH). These tokens are traded freely on exchanges, and their value rises as the market values the potential of the project more highly, even before real token velocity and demand for non speculative stores of value come into play. For example, as the Ethereum network has successfully developed the value of ETH has risen as investors anticipate the emergence of widespread demand for the token generated by systems such as The DAO (which did indeed cause its value to rise). Such tokens are able to reflect the value of progress as it happens and are liquid like publicly traded stocks.

By contrast, although startup businesses can grow rapidly in value they can take many years to generate profits. For example, just consider Tesla, Facebook or even Amazon. The challenge with The DAO model is that profits must be made by investees and shared back by social contract to generate rewards for token holders, and this will likely take years to occur even if the ventures are succeeding. A reasonable suspicion is that most investors in The DAO are seeking returns in a much shorter timespan. The danger then is that once investments are made and the reality that invested funds have been locked up for painfully long and unpredictable periods has dawned the token holders will rush to the exit. The result will be a precipitous fall in the value of the voting tokens on the markets, forcing remaining holders to stay locked in to their investments or crystalize substantial losses.

Of course, our worrying about how and when returns will be created is entirely premature unless the enterprises The DAO invests in eventually generate profits in the first place. This will partly depend upon the ability of The DAO’s voting system to generating good investment decisions in the first place. There are many divergent opinions on whether it will succeed.

These apparently conflicting opinions may all be correct. The current mechanism by which The DAO decides upon proposals is indeed greatly flawed as many in the community suspect — for reasons we will shall examine — and likely to result in poor investment decisions. However, more sophisticated systems are possible. Since The DAO (or its derivatives) are implemented in software they will be able to adopt them with relative ease. Therefore, although the current systems may produce bad decisions, it may be that better systems quickly emerge.

How Payout Proposals are Adopted

The DAO makes decisions as follows. Firstly, someone submits an investment proposal to make a payment to some entity upon which voting token holders can vote (the role of the “Curators” appointed by The DAO is limited to signing that payout addresses are valid from a technical point of view). Each voting token entitles the holder to make a single vote. For a proposal to be accepted and executed, the following conditions must occur:

• The voting period — known as “the debating period” — has expired. This varies by proposal, and lasts between 2 and 8 weeks.
• During the voting period, a minimum total number of votes — known as a “the quorum” — must have been received. The size of the quorum depends on the size of the proposed payout.
• More than half of the votes collected must be in favor.

The quorum calculation algorithm has been chosen so that a relatively small number of votes are required to generate a decision (see http://bit.ly/25aRouC). Payouts of the following sizes can be made with the following quorums and, by implication, minimum proportion of token holders voting “yes”.

payout $10MM — quorum 22.6% — yes 11.4% payout $25MM — quorum 26.4% — yes 13.3% payout $50MM — quorum 32.8% — yes 16.5% payout $90MM — quorum 43.1% — yes 21.6% payout $130MM — quorum 53.3% — yes 26.7%

This default behavior is already focuses the mind — a payout of more than $10MM can be made with the support of just 11.4% of token holders, and a payment of $130MM can be made with the support of just 26.7% of token holders. But the thresholds can be lowered further. To guard against voter apathy, the Curators have been given one more special role, which allows them to drop the thresholds immediately, and then again every two weeks if they wish (see http://bit.ly/1rSSQzT).

If the Curators want today: payout $10MM — quorum 12.6% — yes 6.4% payout $25MM — quorum 16.4% — yes 8.3% payout $50MM — quorum 22.8% — yes 11.5% payout $90MM — quorum 33.1% — yes 16.6% payout $130MM — quorum 43.3% — yes 21.7%

If the Curators want, in another two weeks from today… payout $10MM — quorum 7.6% — yes 3.9% payout $25MM — quorum 11.4% — yes 5.8% payout $50MM — quorum 17.8% — yes 9% payout $90MM — quorum 28.1% — yes 14.2% payout $130MM — quorum 38.3% — yes 19.1%

It is therefore possible that when voting on proposals commences, support of only 4.3% will be needed to trigger a $10MM payout and support of only 19.2% will be needed to trigger a $100MM payout. Conceivably, the Curators might even drop the thresholds further while voting is in progress — although I have no reason to believe that they would choose to do so.

A reddit post discussing these issues is here.

The Lazy, The Active and The Insiders

The thresholds are so low because in practice many voting token holders do not have skills, experience, inclination nor time to investigate investment proposals and will not invest the time to learn how the The DAO’s voting systems work, which only at this very moment are under development. In practice, decisions will be made by three main constituents:

1. Large holders of DAO tokens The top 50 holders control 40% of all tokens https://etherscan.io/token/thedao-token-chart
2. Cryptocurrency exchanges Many users are speculators and will assign their tokens to cryptocurrency exchanges where they can be traded. This makes it possible for the exchanges to vote using their tokens.
3. Crypto activist/enthusiast types Among every community, there is an activist and enthusiast segment — whom will reside on IRC, reddit and other dedicated social media channels — and they can be expected to invest the time to vote.

The low thresholds required and the mix of primary constituents will make the outcome of voting extraordinarily unpredictable, which should encourage those willing to take a chance in hope of receiving a substantial quick payout for a project. Furthermore, it hands enormous advantage to insiders.

The Ticket Ripper, The Ticket Seller

Nick Szabo once proposed a useful anecdote on a crypto newsgroup. He recounted how from the earliest days cinemas have always hired one person to sell the tickets and another person to rip the tickets as customers enter, despite the extra cost in wages. The reason for this approach is simple: it makes it difficult for the seller to give tickets to his friends for free, or sell tickets and keep the receipts for himself. This provides an example of the separation of concerns that occurs many times over in fiduciary systems. One function checks another.

Expediency and rapid evolution has caused a very different arrangement to arise in the case of The DAO. Behind the scenes, the very the firm hoping to receive the largest payout from The DAO is also the project’s designer, creator and primary manager. Inadvertently they have created a moral hazard for themselves: if and when they submit proposals to invest in themselves or their friends, if the proposal passes they could become subject to accusations that the process was rigged in their favor. Indeed, it could easily be. As the evangelists leading the community they hold enormous influence over activists. Furthermore, and perhaps more importantly, as insiders they will be closely linked with large token holders, and as project coordinators they will have established close links with the cryptocurrency exchanges holding a large portion of users’ tokens i.e. they will likely easily be able to muster sufficient votes to pass any proposals they wish.

I am sure that The DAO was architected in good faith, but for these reasons — unpredictability notwithstanding — I expect that very significant initial proposals will be quickly passed in favor of Slock.it and possibly related ventures.

How Bad Decisions Could Manifest

Serial entrepreneurs and experienced venture capitalists learn that creating startup ventures is hard. Wooed by stories of incredible success at statistical outliers like Facebook and Google, the public and indeed inexperienced entrepreneurs and investors are usually highly unrealistic about the challenges new ventures face and the probabilities of succeeding. Talking to people in the community who invested in The DAO, my estimation is few are even aware the venture capital industry in the USA has consistently underperformed the equity markets.

http://bit.ly/24Y6908 (the Kauffman report) https://hbr.org/2013/05/six-myths-about-venture-capitalists (HBR overview)

The challenge is that picking winners is hard, even more so if you are not expert in the area you invest in. Venture capitalists expend enormous energy analyzing deals in depth, bringing in expert opinions, researching entrepreneurs’ backgrounds and looking for support from other experienced investors who will provide an affirmative signal on their own decisions. Top venture capitalists such as Andreesen Horowitz also get exposure to the very best deals in the world by leveraging their brand and profile, and promising to provide extraordinary business support in areas such as hiring.

Many find it difficult to believe that a decentralized investment community will be able to succeed where professional investors fail. I believe it may be possible to succeed, but only with an improved decision making process and in Part III introduce a different mechanism String is developing. Whatever, the following dangers may present themselves to investors in The DAO:

• Adoption of Uninvestable Proposals Ventures unable to raise venture capital may turn to The DAO, but some token holders will not have the experience to see the problems. For example, one of the initial proposals touted by The DAO’s team is funding for the development of an electric car (seehttps://mobotiq.com/). This car does indeed appear to have an innovative design, but serious concerns might be raised about the true total investment required to take a new car manufacturer through to profitability due to the needs of physical R&D and manufacturing, the costs of acquiring required safety certificates, and so on. Furthermore, one might consider that numerous technology giants have already invested billions of dollars in highly competitive new electric cars that will soon hit the road, many of which also include advanced self driving functions. These will be mass produced at low cost and competitively priced while their giant backers battle for market share. The competition will be ferocious. The Google car shown is already constantly patrolling the streets around my office in Mountain View.
• Investing in the Wrong Field Many investors in The DAO are very knowledgable about crypto enterprises, and especially new forms of protocol and decentralized systems. This field is rapidly expanding, and new ventures can grow with low costs because they are based on software, which greatly enhances the chances of success. But these investors will be asked to evaluate opportunities in completely different fields with which they have no experience such as hardware development. This will be demoralizing because they will anyway be bound by unpredictable group decisions when they could have invested directly into more decentralized ventures that took their fancy. A final problem is also serious: if large investments of ether are made into off-chain ventures, they may be quickly converted into fiat currency (in contrast to being held as collateral, or as a legal contingency fund on chain, say). If millions of dollars of ether are dumped on the markets its price will be severely depressed, reducing the value of The DAO’s remaining funds.

Considering the challenges, arguably The DAO will need more advanced decision making processes to achieve returns!

Continue to Part III, Better Crowd Wisdom Using Liquid Democracy…