Ether Investor In Focus: Olaf Carlson-Wee, Polychain Capital

One way to research your investments is to look at what knowledgeable investors are doing. Many investors in the traditional finance community look at what Warren Buffet, the so-called “Oracle of Omaha” is investing in, and listen very closely to advice his gives on how to conduct analysis on his investments before pulling the trigger. Buffett’s investment vehicle, Berkshire Hathaway, has generated 20.8% annual compounded returns from 1965-today, a staggering return, handily beating the S&P500 over the same time period.

With this in mind, it’s useful to look at the investment research process conducted by experienced players in the world of cryptocurrencies.

Below is a video from Youtuber ‘MrYukonC’, who is someone well respected in the Ethereum community, discussing how and who to follow when it comes to crypto investing advice. Around the 6:40 minute mark in the video, he starts talking about how to invest in Ether. It’s well worth a watch, and look at the notes I’ve made under the video.

Later on, I will dive into specific points raised by some investors mentioned by MrYukonC.

MrYukonC on following successful Ether investors

Video notes

6:40 – Buy Ether and Hodl!

Bitcoin over past 8 years has taught us – time is on your side, just hold onto your tokens. Don’t bother trading in and out of your position. Dollar cost average into the position and hold (hodl!). Averaging into a position insulates you somewhat from volatility and is especially useful for younger people.

7:45 – ICOs and Ethereum

Ethereum is here to stay.

ICOs are extremely hot at the moment, so extreme caution is advised. Do lots of research – read the whitepaper, look at the team’s background, ask around. There’s lots of misinformation is out there – people spreading false information.

9:50 – Follow respected people in the cryptocurrency space

Two examples of respected investors are cited:

  • Olaf Carlsson-Wee, founder of Polychain Capital – more details below
  • Fred Wilson, of Union Square Ventures

12:15 – Only purchase ICOs where value flows back into the governance token

– Ethereum, Maker, Augur/REP, handful of others

Looks at Polychain’s account to see what they’re invested in.

Talks about following the strategy of successful people, to follow what they’re doing and contrast it to what he’s doing.

Only buy cryptoassets where some value flows back into the governance token, where the holder of the asset is rewarded with fees/dividends. DGD, Maker, Augur REP are examples of this.

Next, let’s explore some in-depth statements that Olaf-Carlson-Wee has given about how he thinks about his investments, why he believes that funded open protocols are the future of the internet, what he looks for when making investment decisions and how he conducts research.

Podcast: Interview with Olaf Carlson-Wee and Chris Dixon of a16z

Interview link

What is covered in this podcast? In-depth conversation with Carlson-Wee, founder of Polychain Capital, and an early employee of Coinbase, alongside Chris Dixon, who is a well-respected venture capital investor with a16z and financial backer of Polychain.

Carlson-Wee discusses investing investing at the protocol layer, not at the application layer. He and Chris Dixon contrast it to investing in the internet in the mid-90s.

Application layer of the internet = Amazon, Yahoo, Google, eBay – specific user facing applications that regular people use.

Protocol layer of the internet = HTTP, TCP/IP – a layer below the application layer. Not possible to invest in the protocol layer of the internet. Protocol layer is what’s necessary to have a functioning application layer.

Compares buying Bitcoin vs Coinbase. Simply buying Bitcoin would’ve produced similar returns to investing in Coinbase, and isn’t just restricted to venture capitalists (anyone can invest in Bitcoin).

Ether has grown approx x100 in value since launch.

Crowdsale app token launches based on Ethereum’s protocol have raised over $300m to date.

Golem example given of a new marketplace for buying and selling CPU cycles, using Golem network tokens as currency. This is effectively owning equity in a p2p protocol – and betting that it’s popularity will rise.

Often [with new open protocols] there isn’t a company you can invest in – only way to invest is to purchase the token. Contrasted with Linux foundation – which most of the web and smartphones in the world reply upon, yet was/is impossible to invest in, as Linux wasn’t a company. A problem for foundations like Linux is that they must continually raise donations, despite the enormous importance of the work they do. Heartbleed bug – SSL foundation must beg for money despite importance. Open protocols for the web are therefore neglected.

Now there is an opportunity for open protocols to be funded. Creators of new open protocols make lots of money by contrast – effectively Series A (initial venture capital round) for open protocols. Flip side – users of P2P networks are equity holders. Centralised web services Amazon, eBay etc – value doesn’t go to users, only a central entity.

With new token funding model, value flows back to the users, creating extremely strong network effects. Users of centralised services like Facebook and Uber (even power users) are not fanatical about the service, unlike e.g. Bitcoin users.

Current open web protocols: Email – STMP, Internet – HTTP, TCP/IP, HTML.

Very few open protocols developed in the past 20 years, most were developed >20 years ago. This was a Golden era of open protocol development, funded by governments and academia. No capitalist model.

More recently, huge shift to centralisation of power to private companies, dramatic drop off in open protocol development.

RSS was a recent open, social protocol that showed promise in the 2000s. Could’ve rivaled closed social platforms like Twitter and Facebook. Lost to closed platforms. Battle of open v closed in social, closed won. One possible reason is that closed won is they had all the funding and developer talent and better business model.

Open is now developing its own business model, may shift energy and funding in this direction. New funding mechanism – app coins, new infrastructure – blockchain and mining. Open software has existed for some time, open services is new. Blockchain mining is like an open AWS. Public infrastructure, usable for all sorts of things. No central control, can’t be cut off. A developer can develop an app on top of Ethereum secure in the knowledge that no one can pull the rug out from under them.

Growing Linux userbase didn’t increase funding. Need to raise funding indirectly (donations, corporate sponsorship).

Trust. ‘Trust no one’ is baked into blockchains. Assuming that scams and thievery is ongoing and attacking the network constantly. No central authority to shut down transactions. Anyone can participate.

How do you know which projects matter and should get investment? This is Polychain Capitals main task every day. Bitcoin scripting language is very limited. Ethereum has Turing complete language – you can build any application with it. Bitcoin blockchain is only for storage and transactions but Ethereum’s blockchain can do almost anything. Advanced features are possible. Solidity is the language, similar to Javascript.

App specific tokens and new blockchains.

Mentions Tezos – new blockchain. Proof of stake consensus model. Holders of coin act as miners (they validate transactions). Most interesting thing is that the governance pushed to holders of the coin, at protocol level. Decisions about upgrades doesn’t come from closed group of miners or the protocol developers, where it’s opaque, it comes from protocol level, people vote with their coins. Protocol changes based on their decisions – similar to stock holders in a company. Much more democratic. Has a turing complete scripting language.

Example of an application specific token (appcoins) that’s interesting – Maker (symbol: MKR).  Governance and insurance system for a new type of coin that is pegged to the value of the US dollar (called a stablecoin). Because of speculation and trading, most cryptocurrencies are extremely volatile, for many scenarios this makes them unsuitable (escrow, crowdfunding are two examples). This leaves room for a stablecoin that doesn’t change in value. Naive way to achieve this is to have $10m in a bank account and peg the stablecoin to that – however this negates the value that blockchains offer (not decentralised, can be shut down easily, etc). Maker is very ambitious, high chance of failure, but what they are attempting is a system of smart contracts on the ethereum blockchain which create a stablecoin. This coin is pegged to USD. You can then do transactions on the blockchain that you know will hold their value. Anyone in the world can buy Maker, you can get venture capital-type returns if you pick the right coin. Early Amazon investment for example was only available to a small group of venture capitalists, however MKR is open to anyone. MKR is mostly traded on a decentralised exchange (comparable to a decentralised Nasdaq).

Overall theme is bringing capitalism to open source. Why have a hedge fund like Polychain though? Reason: this is an obscure field right now. This is a high risk investment, lots of these assets have declined in value. Those who don’t invest professionally are not recommended to do this.

As an asset class, you can invest in units of protocols – equity ownership in open source projects. When Polychain wished to purchase MKR tokens, there was no legal entity, pure open source project and no bank account. Polychain had to send Ether, in order to receive MKR, no bank account involved. Brand new asset class, totally uncorrelated with equities, bonds, emerging markets, oil, etc.

Polychain’s staff read whitepapers religiously. Bitcoin and Ethereum have famous whitepapers. These define how the new protocol should work. Then they talk to the developer team.

More broadly, they (Polychain) are interested in a stack of technologies that recreate how the current centralised services work, but in a decentralised way. Centralised platforms are built on a massive stack of tech, e.g. DNS, server/client architecture, etc. Needs to be built out in order to have Twitter the protocol, Uber the protocol, and so on in order to compete with centralised companies. So the low level of the internet stack (building blocks) is what they invest in. IPFS project aims to create a decentralised server architecture. Swarm aims to build decentralised apps, similar to Ethereum. Ethereum Name System (ENS), works similarly to DNS, maps “.eth” domain names to smart contract cryptographic addresses. All these underlying technologies need to be in place before application-specific tokens (apptokens) used by normal users can flourish. Polychain also invests in “middleware”, e.g. Golem. Not used by regular users, but by developers. Golem runs on top of Ethereum. Using Golem and Filecoin, developers can build complicated, emergent applications. Too early for viral, end user app tokens. What technologies will enable that?

What are the challenges to be overcome for this vision to become reality? Brand new tech, not just facing market challenges, they’re facing hard technology challenges, sometimes the challenge may be literally impossible to solve. Something like Ethereum was not simply a matter of time, no one knew it was possible. Creating things that didn’t exist, may be totally impossible in some cases. Thousands of developers are needed to build the decentralised web stack, and because of Polychain and those like them, more funding is becoming available. Matter of time before some of these projects become the next era of the internet.

What about the problems of coordination, are there challenges unique to these open source projects? In open source projects, the arguments are in the open, in private companies it happens behind closed doors. Conflict resolution is build into blockchain protocols – miners can vote on Bitcoin’s blockchain via their hashing power. Tezos is an alternative governance mechanism allows one coin=one vote democracy.

Lots of competition as this is a growing asset class.

Long term vision is where the web is owned by users of the web. Privacy, data mining, and pseudo-monopolies are huge problems today. Wave of technologies are emerging where the pendulum could swing in the opposite direction. You own your data, you decide what to share with services, which will be p2p services owned by the users. All the wealth generated by VCs and entrepreneurs of centralised companies, will get pushed back to everyday investors and most of the value will go to early adopters and users of these platforms and protocols instead of owners of web platforms.

Emergent behaviour. Easy to compared the early days of cryptoassets to centralised web platforms – it’s what we know. Internet wasn’t a digital library, it was much more that was impossible to predict in advance. Therefore cryptoassets will allow things currently unimaginable, just like the internet did. Biggest things that will emerge are ont currently imaginable.

Podcast 2: Interview with Olaf Carlson-Wee on Forbes Unchained Podcast

Interview link

1:30 Polychain is a hedge fund that invests exclusively in digital assets. Invests in protocols not companies. Invests in things made scarce through the blockchain.

Polychain is a hedge fund that invests exclusively in digital assets. Invests in protocols not companies. Invests in things made scarce through the blockchain.

Investing in protocols – Satashosi Nakamoto solved Byzantine Generals problem, thereby creating digital scarcity. With Bitcoin, funds are definitely transferred – solving the double spend problem. Not possible to copy and paste Bitcoin. Created a whole new area of innovation. New platforms are emerging to allow almost anyone to launch tokens/scarce digital assets. That’s what Ethereum offers.

3:50 Security of a blockchain is 1:1 with it’s market capitalisation. When launching a new blockchain, it’s vulnerable in early stages. Therefore launching on top of another blockchain that has solved these problems is very valuable.

4:30 Terminology used- Cryptoassets, Blockchain-based assets. Trying to shift the perception that Bitcoin/blockchain technology is limited to money or currency- this is not the full picture. Ethereum offers much more than value creation and transfer. Offers the ability to create arbitrarily complex pieces of software on the blockchain.

5:45 Examples of things that Ethereum enables. Ethereum has a Turing complete scripting language. Bitcoin’s language is deliberately limited, only possible to store and transfer value. Ethereum’s scripting language means you can build much more complicated financial contracts. Known as smart contracts. Gives example of playing blackjack against a smart contract. Taking actions in the game (in blackjack lingo – standing, hitting, shuffle) means interacting with the state of that smart contract. Possible to win (and lose) Ether playing against this smart contract.

8:30 Polychain Capital Details. Currently around $15m assets under management. Investors – Andressen Horowitz, Union Square Ventures, and a few others. Launched the fund to invest in blockchain based assets. Offers exposure to broad based digital assets, without the usual way of simply buying the asset directly.

9:30 Carlson-Wee’s background in the blockchain world. How he first heard about Bitcoin – June 2011, friend told him about Silk Road, the darknet marketplace. Article casually mentioned Bitcoin. Did lots of research on Bitcoin. Started buying Bitcoin. Wrote undergraduate thesis on cryptocurrency. Very little academic studies at the time on Bitcoin. At the time, the BTC price drop from $17 to around $2. Articles at the time predicted the fall of Bitcoin, a computer science experiment that failed. Held onto his faith in Bitcoin, and adoption rose, and so did the price. Knew immediately that Bitcoin was the most important technology that he’d ever heard about. Whitepaper for Bitcoin was released just two years prior. Felt like he was reading whitepaper for the internet. Felt lucky to get in on the technology so early. Programmatic finance – possible to run the financial system through software, rather than centralised clearing houses, banks, etc. Just like the internet has disintermediated many industries, crytocurrencies will be automated through software.

I knew immediately that Bitcoin was the most important technology that I’d ever heard about. Whitepaper for Bitcoin was released just two years prior. It felt like I was reading the whitepaper for the internet.

14:50 How he went from thesis to being hired at Coinbase. Bitcoin industry was non-existent at the time. Coinbase launched, he was the 30th user. He was using many of the services that were being launched at the time to test them. Coinbase had a killer feature – being able to buy Bitcoin on the internet! Easy to imagine that it should be easy, but a big problem is that if you pay with Paypal or credit card for Bitcoin, you can do a chargeback, and the seller doesn’t get the bitcoin back. Traditional payments are reversible, Bitcoin isn’t. Fraud therefore, is a massive problem for sellers of Bitcoin. Coinbase offered Bitcoin purchases via bank transfer. Very hard problem to solve. Prior to that, Carlson-Wee made Bitcoin purchases in cash. Describes the awkwardness of buying Bitcoin with bank transfers. Saw Coinbase growing and growing, was purchasing Bitcoin constantly, saw the potential in Coinbase. Cold emailed jobs@coinbase, was willing to work in any position in Coinbase. Cofounder replied quickly.

20:30 Jobs at Coinbase. Worked customer support. Did this alone until they had 250,000 users (when he started it was 50,000 users). Automated a lot of the support requests, matching keywords. This solved a lot of routine tickets. Next step, he hired a distributed team. Difficult to hire, needed to fix bugs quickly as a lot of money was on the line. Customer support tickets were first indicator of things going wrong. Funds flows between BTC addresses and bank accounts. Much higher requirements than typical customer support job.

24:45 Created the “Bitcoin SAT” – quiz with complicated Bitcoin questions. Used to screen potential hires to test ability. Posted it on the main BTC forums. If you get perfect score, you will get an interview. Got a huge response. 250 took the test, lots actually did well. Started doing 4-5 interviews per day. Very high calibre of interviewees. Paid all the customer support, all over the world, paid in Bitcoin. Team was working in 12 different countries, so it was easier to pay them in Bitcoin rather than dealing with multiple currencies.

28:35 – Next step after customer support. Head of risk. Dealing with fraud and user account issues. People buying bitcoin with stolen identities. Coinbase is on the hook for the loss. Talks about anti-fraud systems put in place. Was given leeway to work on whatever he wanted, started looking into Ethereum.

30:30 First experimental token launches around this time (late 2015-early 2016). Series A style funding for open source, P2P protocols, a completely unprecedented phenomenon. Needed to leave Coinbase to become more actively involved in Ethereum and token launches.

31:45 What’s so different about these token launches, versus earlier tweaks to the open source Bitcoin code, launched as a new digital currency? 1. Now it’s much easier, Ethereum makes it much easier. 2. People raising money to fund the protocol – effects normally seen in venture rounds are now open. Anyone can participate. Funding an open source P2P project that funders want to see exist. Main effect is – monetised p2p networks. P2P networks historically never had a monetary layer attached, e.g. Torrents, Tor protocol. These never added payments, they were run altruistically. No-one is doing it for money. Now with digital assets, you get asymmetrical P2P networks. One side is extracting, the other contributing. You can have P2P marketplace. Golem is an example, a P2P market for computation. Rent out CPU and GPU cycles to someone who is training a machine learning algorithm. P2P alternatives to centralised hubs, e.g. Microsoft Azure or Amazon web services.

35:00 Why create new token instead of using Ether for payments? Mostly, not for hard tech reasons, but for game theoretic reasons. In the Golem example, allows GNT (Golem Network Token) holders to be exposed to growth of the Golem network. Also tied to governance – used for voting on the network. As an investor, you are exposed to the growth of the Golem network, not broadly Ethereum. Founders of the protocol have equity incentives – where if the network becomes very popular, it’s possible to see a very large rise in the value (e.g. x100-x1000) of the tokens. So issuing own tokens aligns incentives better, and provides a better governance mechanism.

37:25 – By issuing own tokens, you can create funding out of thin air, sidestepping Venture Capitalists, instead doing a crowdsale and raising several million dollars. Because of how appealing this sounds – a lot of less legitimate tokens/scams may appear. How does Polychain evaluate which tokens are worth investing in? Totally permissionless innovation. Anyone can create a token. Also means a lot of noise, his job is to find the signal in the noise. They read lots of whitepapers religiously, they look at the Github code repository to check the codebase, as well as the Github forks and stars (indicators of interest from other coders with the open source codebase, i.e. the developer ecosystem), as well as speak to the founding team, particularly the technical architects on the team to make sure everything makes sense. Carlson-Wee doesn’t have a computer science background, so he gets others on the team to do that part of the research. Forks = copy the code and make some tweaks.

40:50 Importance of developer interest in a project. What they’re (Polychain) investing in currently are protocols (or “middleware”), Golem being a good example. Golem won’t be used by “regular people”, like a user-facing app. It’s a developer tool, so that they can build user facing apps. The users in Golem’s example, are developers. Ethereum is also a tool for developers. Therefore, measuring the developer ecosystem is measuring the success of the project. If a lot of developers are showing interest, and trying to build projects on top = very positive signal.

Measuring the developer ecosystem is measuring the success of the project.

41:55 What are the main reasons to pass on an investment opportunity? Many! A lot of people wish to create a token to raise money, but the token doesn’t make sense for the project they’re building. In general – if you are building a project with network effects, then a token can make sense. If you’re not building a project with network effects, then a token probably doesn’t make sense. This has remained true for many projects. As a buyer of a token without network effects, the project could succeed, but the value of the token might not increase at all. When buying tokens, Polychain must create returns for it’s investors, therefore they need to invest in things where the growth of the network will be 1:1 with the token price increase. That’s the number one red flag to not invest.

If you are building a project with network effects, then a token can make sense. If you’re not building a project with network effects, then a token probably doesn’t make sense.

43:45 Apart from Golem, what other digital assets are they excited about currently? Ethereum and Golem are application-specific tokens. Golem isn’t general purpose money, only for the computational marketplace. Another token, not Ethereum-based but on another blockchain is Tezos. It is like Ethereum, Turing complete language, capable of allowing developers to build smart contracts, but has different qualities. The language is capable of being formally verified, meaning that you can mathematically prove that the contract will do what it’s intended to do. This means it’s easier to write a bug free piece of software. When writing financial contracts, bug free code is very important, e.g. avoiding a situation like the DAO hack, where someone exploited a bug to steal millions of dollars in funds. Additionally, Tezos is proof of stake blockchain. 1 coin = 1 vote, on consensus. Therefore token holders take care of the consensus mechanism, instead of miners like with bitcoin. Lastly, it has protocol-level voting on decision making. Governance has been historically a difficult problem in the blockchain world. With Bitcoin, it has manifested as a multi year debate around how to scale the protocol, with large disagreements between the various factions, and no clear governance on how to decide on what to do. This has resulted in an analysis paralysis, meaning neither side wins. In Ethereum, a decision was made to fork the protocol after the DAO hack, and un-do what the hacker did via a fork in the blockchain, which was a controversial decision. A vocal minority opposed this, and there was no good mechanism to determine how to move forward. With Tezos, 1 coin = 1 vote, so you have votes on how to proceed, and the protocol will automatically upgrade/fork based on votes. So you have built in governance at the protocol level, not based on vague relationships between developers, miners, users, exchanges… but codified governance. Will be interesting to watch over the next year or so how it develops.

You have built in governance at the protocol level, not based on vague relationships between developers, miners, users, exchanges… but codified governance.

47:10 Does he worry about legal/regulatory issues affecting his investments? Yes, a little. Regulatory risk is there, unclear regulation. That said, open source, and P2P sectors have proven to be extremely resilient. For example, with torrents and file sharing, they’ve been extremely resilient to attempts from any particular geographic region to squash that activity. This is a global phenonema, so there is no clear legal jurisdiction over these projects. All happening on the internet, and the digital assets held by the fund are not domiciled in any particular geographic region. Really exist on the internet. Truly global, internet based phenomenon. Legal frameworks are based on geographies in the physical world, not on relationship over the internet. Therefore the legal situation is complicated. If any country makes laws affect this area, it’s unclear what that applies to. If a project is online, and no developers are US-based, likely outside US jurisdiction.

Legal frameworks are based on geographies in the physical world, not on relationship over the internet. Therefore the legal situation is complicated. If any country makes laws affect this area, it’s unclear what that applies to.

49:05 Polychain holdings – how do they keep the private keys secure? Industry standard, offline (cold storage) wallet. Get an offline computer, that has never touched the internet, and never will, and generate keys on that computer. Cryptographic keys that store the cryptocurrency. Then you can create backups of these keys- on flash drives, and on physical pieces of paper, which never touches the internet. You can secure these in safe locations, like banks, safe deposit boxes. When you’ve created a cold storage wallet, you can then send cryptocurrencies to this wallet, completely securely. You can, counter-intuitively, send cryptocurrency from an online exchange to a piece of paper. What is on the piece of paper is access to the wallet – which is stored on the ledger, on the internet. Very easy to move from an online environment, to an air-gapped (offline) environment. Means that an attack targeting the cryptocurrency needs to happen 100% offline. This is the most important step when securing your cryptocurrency – to move to offline storage.

51:05. Background to how Polychain got large, well-established Venture Capital firms to invest in the hedge fund. Very unusual for a VC fund to invest in a hedge fund. This is a very unique asset class. Bitcoin companies raised around $1.4bn in VC financing, Ethereum based companies have raised a negligible amount. By contrast, Ethereum digital assets, built on the platform, have raised around $300m – order of magnitude more than what’s been raised by private companies with VC financing. Lots of activity, and almost none of it VC funds. Gaining access to these digital assets and protocol based tokens means you have to buy directly into the tokens, or invest in something like Polychain. For a traditional venture firm, explaining to your investors that you will purchase cryptographic tokens, is very hard! Investing in another fund, while hard, makes more sense. Andressen Horowitz and Union Square Ventures had great faith in Polychain, and therefore decided to invest to gain exposure. They also know that monitoring the space, reading the whitepapers, talking to developers, and understanding the ecosystem, is a full time job. They put that on Polychain, to avoid having a whole team of people dedicated to monitoring this area. The skillset involved when conducting the research is quite specialised.

53:50 Background on how Carlson-Wee was making and spending all of his money in cryptocurrency. Creating Polychain has complicated the “living on Bitcoin” experiment that he was doing for several years. He put 100% of his crypto holdings into the fund, so he doesn’t have cryptocurrency sitting around in personal wallets.


Disclosure: I have invested in the Tezos ICO and hodl ETH.

Now read: Should I invest in Ethereum?