We're just back from Berlin, where this week we had the pleasure of speaking at Project A's portfolio day.
Aside from the usual introduction to tokens and cryptocurrencies, we focused our talk on describing what we've been telling people privately, and on here, for a while: we're not big fans of ICOs.
Fortunately we're seeing data that suggests the trend of anyone being able to raise whatever they wanted for any kind of projects is over, but the frothiness is still rampant.
The main points that make us be short on ICOs are:
1. Too many people participating
This is the most controversial, so just putting it up top as usual.
We're all for the democratization of both access to capital and deployment of capital, so a better phrasing would be: too many people participating in the same ICOs.
The more people participate then the less potential upside there will be for each single participant. Unless the upside is infinite, every new participant will push the price of an asset a bit higher, up to the point where it doesn't make any more sense compared with the risk.
The other issue is that there are many different types of people participating. In any given ICO you'll have short-term speculators, long-term holders and actual prospective users. This entangles the price of the token with many different variables, making it hard on the latter two categories.
2. Risk-reward is completely out of whack, as most projects are just vaporware
What I'm saying here is that I'm happy to invest in cool vaporware, but not at a 9 figure valuation.
With the influx of less sophisticated investors, risk is being greatly discounted and any potential gains will be dwarfed by the losses on the projects that don't work out.
Let's remember Ethereum had something like a ~$20M network val when it did its fundraise. That was still pretty steep compared to the state of the project, but also orders of magnitude less than what we're seeing today.
3. Setting in stone crypto-economic decisions
I've tweeted about this quite a lot. Cryptoeconomics is hard. It's new and no one really knows what they're doing. Many projects are setting in stone their crypto designs in early whitepapers written by consultants just with the goal of raising money. Building companies is hard, and building decentralized incentive ecosystems is even harder. Just thinking that you can get something so complex right at such an early stage should make people run away.
4. Potential adverse selection of teams
We've seen this very clearly. The best teams we're talking with are pondering the ICO decision VERY, VERY deeply. And if there's a way to avoid it, they will - at least until the software is ready, cryptoeco decisions are made and people can use it.
The market is attracting all sorts of participants that are following the bling of quick and massive non-dilutive raises - but nothing has changed: building something of value is still super hard. Good teams and entrepreneurs understand this, and understand that an ICO could be more of a liability than an asset.
The main point here I think is that we like to see teams that *care* about other people's money. Most teams who ICO treat the capital as their own war chest and "free money". We think the best entrepreneurs realize the moral and reputational responsibility of being entrusted with capital by other people and thus naturally try to get the least amount that they need.
But we also get the catch-22. If everyone around you is raising tens of millions, who's the fool?
If you're not gonna do an ICO for your new Mars colony, then someone else will - and people will give them money. So now you'll have to compete with someone that has a lot more cash than you.
I unfortunately don't have a great solution here.
We could add many more reasons to be fearful of ICOs, such as the need for more securitization infrastructure, governance oversight, investor communication, and such - but in any case, nothing here changes our deep belief that crypto tokens are here to stay.
Tokens are a fundamental shift in how value is assigned and transferred in ecosystems.
There are so many transactions today inside and between ecosystems that are simply not valued, or implicitly valued. Having fixed-supply, immutable ledgers keep track of these transactions could be a game changer for some specific types of applications.
Being able to apply monetary policy and fiscal policy to an application and its users could give many more tools to developers compared to the minimal UX tools we have today.
We're seeing this in action in thoughtful ways more and more. One of the best examples of the past week is Refind, and its Relevance Coin.
Refind grasped the potential of tokens and decided to "give away" 1 billion tokens to its users, based on a few simple potential actions to start (eg. inviting users, sharing on social media etc.). They then promised to issue the tokens as ERC20 or similar crypto tokens when they allocate the full 1 billion supply and then start to buy them back with a percentage of their revenue.
This is fascinating. Developers figured this out early by giving away small rewards or even just worthless karma points or something, but now you can invest your time (and money potentially on futures markets) with the expectation of financial return in the future.
Refind's implementation is still massively flawed: - you are trusting them to issue the tokens - you are trusting them that they will actually buy them back, something they will have absolutely 0 incentive to do - you are also wagering that they will actually get to 10m users
but it's a great experimentation to see. The most interesting aspect to think about in my mind is the buy back mechanic.
Say we believe Refind will use 10% of its revenue to buy it back. The easy thing to do would be calculate their revenues and see how much money would go to "shareholders". But here it's not a dividend. They will actually buy the tokens on the market AND burn them. So this would mean that every month or so, there would be a growing price pressure on the token from them, but *also* from other market participants that want to gain from the gaining popularity of the platform and their (hopefully) growing pressure on the price. And, a constantly decreasing supply.
So essentially, a Berkshire-like constant stock buy-back.
I'd need to spend a bit more time thinking what the decision tree for a token holder would look like, but I think those are some early signs of the cool things that crypto will enable.
Take note everyone: you can use tokens in smart(er) ways without selling them.
The talk of the town the past week has been Basecoin.
We are glad that the Basecoin project gets finally unveiled and people can comment, as we have spoken to the founder in early September and couldn't come to a massive enthusiasm about the project. Something other famous and reputable investors were able to do. This has left us with the doubt of being stupid, or others being overly-excitable. Nothing new in venture land.
I'll preface this by saying that I've never really felt a need for a stable coin. In the US there's Venmo, in Europe we have free SEPA transfers with just an IBAN as well as many Venmo-like startups - so the p2p payment problem isn't that major. But I've slowly come to change my mind, and have recently developed the thesis of the need of moving money "into" fiat, but without having to interact with the traditional banking system.
The idea that Basecoin proposes is that of a coin peggable to any asset or basket of goods, whose price is adjusted by the algorithmic expansion or contraction of the monetary supply on the blockchain. Kind of an algorithmic Central Bank.
The algorithmic execution of the peg involves the existence two other coins, Base Shares and Base Bonds, according to the need of contracting or expanding supply. If you want to dig deeper you can have a go at the whitepaper yourself.
The questions we asked during our chat were mostly around what would happen in systemic, exceptional price movements. I don't think we were either able to receive great answers or either understand them. But the feeling has always been that this is a very nice model in somewhat stable conditions but that there was no core, structural mechanism to prevent collapse in an exceptional event.
The whitepaper doesn't do a great job at this, delegating all the discussions about robustness to a subsequent paper. What they do is create an artificial bottom price of 0.10 Basecoins for 1 future Basecoin, and they say their simulations never hit that.
For all the reasons why this will never work it's worth reading Preston Byrne's post on the subject. He seems to have the same doubts we had, but with much more conviction and bitterness at yet another doomed stablecoin project.
For a more positive take read Coindesk piece (who are owned by an investor in Basecoin).
In any case, we do find the effort fascinating. It's a clear sign of the scale of innovation that is going on in the crypto currency space, and that's the reason we're excited about being here.
They are preparing for a pre-sale and potentially an ICO soon. As with many projects, we would have hoped the funds raised in this round lasted for a bit longer to get to market with much more progress.
After Emin Gün Sirer famous piece "Bancor is Flawed", which demonstrated, only a couple of days post their $153 million ICO, how miners could front run transactions, a Google software engineer found another major vulnerability: even non-miners can front run on the Bancor exchange. The post goes through the code.
It turns out Ivan has been working closely with the Bancor team on a fix to this vulnerability and only published the research ex-post.
It's always awesome to see white hat hackers in action, but also hopefully illustrative to folks on why you don't hand over $150M+ to any project at this stage, no matter the potential.
This is a super cool article on Token-Curated Registries, which are a way to keep decentralized white-lists. We love articles that explore the cutting cutting edge of crypto-economic thinking, and TCRs are right there.
Hopefully people start realizing how cool this stuff is compared to ICOs soon. We'll be here waiting.
The concept like in many other crypto-systems is to have staking mechanics to incentivize people to self-curate and self-organize (in this case registries). You post a stake, and if you're challenged and you don't fit you lose it.
The implications are obviously many. A lot are covered in the article, but I feel there should be much more thinking giving to self-policing protocols.
Elad explores one of my favorite topics: the problematics of token funding and what we can learn and port from the world of startups and VCs.
It seems that my thinking is very very similar to Elad's, so I'll just refer to his post and thank him for writing it.
The main takeaway for me is: Tokens are changing the game, but not for every single type of project. We still have to build companies and align incentives. The likelihood that your company is the one that benefits from being funded 100% by token sales is low.
We spoke about atomic swaps before, and the developments are continuing fast. Ethereum contracts do make them quite easy at least theoretically, and now there's an open source one that anyone can implement and continue developing.
Decentralized trading and cross-chain functionalities are coming fast.
This is a fun list of predictions from people in the space.
Laura had kindly asked me to provide some input but I failed miserably to deliver with ample time. I'm blaming this on you. The reality is that I find it extremely difficult to explain what is going to happen, because I don't know. It's fun to say "oh the future will look like this" but 8 years ago nobody could have predicted what we're doing today with blockchains and tokens.
I prefer to be close to the people inventing the future and ideally try to fund them.
Anyways, fun surprise tidbit at the end of the article: we learn that Naval is not CEO at AngelList anymore.
It turns out UNICEF has an active $12.6 million blockchain focused venture fund.
It also turns out they are running the first public ethereum node managed within the United Nations and that they started accepting donations in ETH to their ethereum multi-sig address as a test last August. It's great to see blockchain being adopted in unlikely places.
As to doing an ICO, any money that goes to charity is a great use of it, but we also hope there will be a real, well thought-through use for a token, so that it can start to bring even more innovation and capital to this space too.
CENTRE comes from Circle, one of the first Bitcoin companies around.
This appears to be a massively ambitious project of proving a set of standard protocols for payment services that want to use the blockchain to send fiat. These protocols include support for transmitting fiat money over blockchain implementations, a service provider mechanism to support trust and identity decisions, rules for payment settlement and reversals, and the secure exchange of KYC/AML-related information to meet compliance obligations.
Not that taxes are cool in any shape or form, but this product coming out in January aims to be the Turbotax of crypto.
Could be handy for all the profit takers.
👮 This week in regulation
The usual sequence of central banks and regulators issuing statements regarding crypto currencies and ICOs. It's typically a little bit dry, so unless it is something worth commenting on, we'll just post the links for the record: