Notes on Placeholder blog posts

Written in: December 2019 Notes

Last updated on: December 23, 2019

Topics: Crypto

As part of my research to develop a sound crypto thesis, I’m reading (or re-reading) all of the Placeholder blog posts and taking notes…

Investment Thesis - Feb 2018
  • Crypto starting in 2010 is equivalent to hardware dominated by IMB in the 50s/60s, software dominated by Microsoft in the 70s/80s, and the internet dominated by Google/Facebook in the 90s & 00s. open-sourced/decentralized data will define the next 1-2 decades. Most of the companies from the internet era make money off having monopolies on certain cuts of data collection. The next phase will see people own their own data.
  • Hardware, software, and the internet are all open-sourced now. Only data still isn’t open-sourced. It’s a natural cycle for new innovation to democratize what’s been monopolized in the previous cycle
  • Crypto incentivizes companies to maximize for use/utility of their product. Right now incentives are towards profits, and the two aren’t always correlated
  • Can achieve unprecedented levels of scale and operate on a truly global level.
  • Investment strategy: “We fund the development of decentralized information networks coordinated by a scarce cryptoasset – or token – native to the protocol. Our thesis is that decentralization and standardization at the data layer of the internet is collapsing the production costs of information networks, eliminating data monopolies and creating a new wave of innovation. Cryptonetworks accomplish this by replacing expensive, centralized coordination (e.g. PayPal) with universal financial incentives (e.g. Bitcoin). These networks introduce a new, natively digital asset class which shifts value away from equity in companies to tokens in decentralized networks”
  • They segment the market into three layers
    • Infrastructure protocols: low-level, general blockchains focusing on security, scale, and consensus (Bitcoin/Ethereum)
    • Decentralized applications: built on top of infrastructure protocols, but tie it all together on top of their own token to fill a niche. Compete on the basis of community, governance, and cryptoeconomics
    • User interfaces: add ability for every-day user to interact with dapps. Usually don’t have their own token, but can add other functionality on top. Employ traditional business models for advertising, subscriptions, etc.
  • Protocols will never die so long as at least one node is still running. It’s not up to one company to decided to pull the plug or on the direction of the token/dapp
  • Focus on:
    • Decentralized information networks
    • That are coordinated by a scarce token
    • Which appreciates in demand as user demand for the service grows
    • Avoid equity businesses with traditional revenue-based business models, centralized services that use a token purely as a fundraising mechanism or payment method, private blockchains, and other implementations that would not be described as “open.”
  • There is no long-term defensibility in the code or data of these companies. They can always be forked. So you have to create community and trust with good governance. Founders are very important
  • Carefully consider the cryptoeconomics
    • “Protocols which shift more value towards the users are more likely to succeed than those which over-concentrate value in a small group of developers or investors (including ourselves)”
    • Crypto networks are less like companies and more like small emerging countries that export one good
      • The consensus protocol is the constitution
      • The community is the constituency (miners are supply, users are demand)
      • Core developers are the executive branch (they are in control, but require sign off from constituency to make radical changes)
      • The token is the internal currency
      • The investors underwrite the currency
    • The same metrics that are good at evaluating a country apply here. Good governance, sound monetary policy, low corruption, low inequality, productivity trends and so on.
Cryptoasset valuations - Sep 2017
  • First example is using the velocity of money supply equation (MV=PQ). As demand for a dapp increases, the demand for the tokens will increase and the value of the coins in circulation will go up.
  • Usually solve for M=PQ/V
    • P is the price of the service being provided (for filecoin, the comparable cloud storage per unit)
    • Q is the quantity of the service in the market
    • So PQ is kinda like the GDP of the small economy
    • M is the market cap of the coin
    • V is the velocity of money (how many times the average coin is spent in a year)
    • Should ignore transactions like people moving coins between wallets -> not real value or exchange
Decred Investment Thesis
  • Hybrid proof-of-stake/proof-of-work. Miners mine blocks similar to bitcoin, only randomly selected users based on stake are selected to verify that the mining is valid. This keeps things fair and mitigates risk of 51% attack or large miners doing all the work
  • Rewards are split 60/30/10 -> miners/stakers/development team. This allows miners to verify transactions, stakers to verify miners, and the development team to continuously fund themselves
  • Almost 50% of DCR is staked
  • Network hashrate exploded
  • Good governance and founder (who also put a lot of his own money into projects)
A brief study of cryptonetwork forks
  • When a coin chain-forks (copies the ledger and keeps the codebase mostly original), users who owned in the original chain get the same amount of coins in the new currency (though not value) automatically (Ethereum Classic).
  • When a coin codebase-forks, they change the way the network operates significantly and start an entirely new currency that just happens to be similar to and existing currency (ZCash and Litecoin)
  • It was originally thought that forks would be dilutive to users of a currency, but it seems that forks actually create new user bases. Users remain loyal to the original currency while miners flipflop to wherever they can make the best return. Core developers also remain loyal (not surprisingly)
  • Forked projects often get abandoned by devs over time. Maybe it comes from a place of spite, which doesn’t last as long as passion
  • NVT ratio seems super useful -> market cap (volume) divided by daily USD volume of transactions. It’s a velocity of money measurement
  • Metrics for daily active users/addresses is also helpful
  • Forks usually trade at a higher valuation (higher NVT) to their parents. Could be because people just let those coins sit (less selling pressure), or that they view the coins as insurance in the parent chain (again, less sell pressure)
Resource Distribution and Power Dynamics in Decentralized Networks
  • In sociology, the term field denotes a structured social and symbolic setting in which individuals and groups acquire positions and act, and in which systems of meaning, institutions and hierarchies are formed, maintained and challenged.
  • Fields are not neutral, open and free marketplaces — at any given time, there are norms, limitations, inertia, and various structural forces
  • Decentralized networks can be thought of as fields
  • Fields tend to have similar properties:
    • Fields form when a large enough social group is established
    • Competition and cooperation over field-specific resources exists
    • Unequal distribution of these resources
    • Tendency of dominate power to struggle for consolidation and conservatism; tendency of subordinate positions to strive for change and disruption
    • Tendency of dominant group to justify unequal resources under a common belief
    • Relative position in the field has an impact on individuals (whether they recognize it or not)
    • Tendency to have economically and culturally dominant poles
  • We need to look at this definition of field theory and apply it to decentralized networks to determine if they can remain in balance over time and what sorts of imbalances might arise.
Cryptoassets: Flow & Reflexivity
  • 2018 saw a “crypto tax crisis” as people were forced to sell assets to pay unexpected tax bills
  • These “flows” can severely affect illiquid crypto markets
  • These flows can set of “reflexivity,” which is when investors create a self-fulfilling prophecy with an internal feedback loop.
  • Crypto has higher reflexivity than more mature capital markets. This is likey because it’s still early-stage tech and people are speculating
  • The liquid nature of assets also amplifies the problem
  • Might see dips in the market centered around March/April that coincides with tax payments. Also December, just like stocks, as people rebalance portfolio and sell off positions before the end of the year.
  • Reflexivity should dampen over time as markets mature
Productive Capital in Cryptonetworks
  • Two ways to invest in crypto assets—minting to supply-side who employ productive capital, and selling to investors who commit investment capital
  • Supply side invests into productive capital, but then sells off what they mint to investors. Investors are second in line
  • Selling to investors before minting places investors before sell siders
  • Minting models can fall flat when there’s no investor interest at the end of the tunnel
  • Likewise, in investment-capital-driven markets (where investors fund development of protocol), investors are a given, but they need to focus on the protocol to make sure supply-siders eventually come around to mint and entire demand-siders
  • Tendency in last few years for companies to over-prioritize investment-capital-driven markets with risk of neglecting supply-siders, who aren’ needed now, but are essential for a healthy network
Crypto networks are not companies
  • Crypto markets are like economies, not companies
  • Only services that thrive off decentralization will do well under crypto architecture
Maker investment thesis
  • First 100% software-based, community owned lending platform
  • Byproduct of loan is Dai, a stablecoin
  • Borrowers lock ETH into a collateralized dept position (CDP), then issue themselves a loan denominated in dai.
  • The amount of dai they can borrow depends on loan parameters set by MKR holders, but it can’t exceed 2/3 the amount of locked collateral
  • If loan is past due or if the collateral dips below 150%, keepers in the network can enforce liquidation of the locked assets to pay down some or all of the debt.
  • Supply of dai expands as users pay down loans (since they mint dai by paying them off)
  • The locked collateral incentives borrowers to use the money they borrow wisely. And since everything is decentralized and not run by a company, interest rates are significantly lower (0.5-2.5%)
  • The interest paid on a loan (or stability fee) must be paid in MKR, which is subsequently destroyed, making it deflationary
  • While MKR holders stand to gain from the deflation as the supply of MKR decreases, they also stand to lose if debts go bad. If at any point a loan goes over 100% collateralization, new MKR is minted and sold to pay the loan down to below 100%.
  • It’s up to MKR holders to enforce rules to make sure loans are always properly collateralized
  • Dai is already being used by Ripio, Wyre, Compound and Nexo
  • The largest benefit over the current global debt system is transparency. Anybody can look at collateralization and “inspect the books”
  • The people earning interest in MKR are also incentivized to manage risk properly. And if the system fails, there’s no government bailouts. All participants are harmed equally.
  • This system also provides cheap credit to anyone, anywhere.
The defensibility of middleware protocols
  • Middleware or service-layer protocols sit between infrastructure protocols (layer 1) and operate in a niche they can mostly dominate, with various user interfaces interacting with the protocol.
  • Financial-service protocols that Placeholder has invested in include 0x, Erasure, MakerDAO, and UMA, while Aragon is our main social-service protocol to date, and technological-service protocols that we work with include CacheCash, Filecoin, FOAM, and Zeppelin. All of these protocols have originated on Ethereum, but we believe interoperability of state and value—the promise of a Cosmos, Polkadot, and Ethereum 2.0 future—will allow these protocols to become horizontally defensible starting from Ethereum’s base.
  • Protocols whose reliability, security, speed, liquidity, or coverage scales with the size of the asset base and nodes supporting it, stand to do well in an interoperable world.
Value capture and quantification: cryptocapital and cryptocommodities
  • Stake-based coins are more capital assets than traditional crypto coins, which are commodity assets. This means you can’t us MV=PQ to value them.
  • Any cryptonetwork that requires ownership of the native cryptoasset to gain access to a recurring value stream generated by the network, thereby creates a capital asset as opposed to a commodity.
  • Any purely proof-of-work asset can be considered a cryptocommodity, and MV = PQ remains our best bet at pricing such assets.
  • We traditionally think of something of value as being something that creates cash flow, but crypto needs new metrics
  • Marginal cost of production acts as a floor -> ie the energy cost to mine bitcoin is the floor because once it dips below, more miners will jump on. If this doesn’t hold for long periods of time, the network might not be healthy.
  • The years post-halving tend to be strongest years because you have the same number of people competing to mine half as many coins -> they become more scarce
  • It took 300 years from Graham and Dodd to figure out how to value stock markets. We are only 10 years into crypto, so it might take a while before we hash out a solid valuation framework
Protocols as minimally extractive coordinators
  • Crypto networks are not businesses, they are exchanges
  • Protocols need to be minimally extractive in order to maximize utility, while businesses need to be maximally extractive to make a profit.
  • The less extractive the exchange is, the more that exchange will be used (and others like it)
  • Unnecessary extraction is a tax
  • Crypto can be minimally extractive, while still creating a lot of value because of the scale they’ll be able to operate at globally
  • Supply-siders (miners, stakers, etc.) are businesses -> the have income statements and use the crypto networks to run their business
  • Distributors build apps on top of networks and connect them to users. They are also businesses.
  • Consumers will have the option to pay the protocol, or the sugar-coated distributor (more expensive, but easier to use) -> most will probably choose the latter
  • Crypto is the incentive layer that ties together the supplier, distributor, and consumer
Fire before growth: the likely fate of Ethereum killers
  • Many EKs will debut, some of which with billion dollar network valuations
  • Expect many EKs to start off with extreme downward pressure as initial pushback for high valuation
  • EKs are entering a bear/sideways market looking for blood, not the time to go public with high valuations
  • “In bubbles, everything is evaluated in relation to other parts of the bubble, not to the world at large” -Albert Wenger
  • Lots of EKs emerging at once will spread developers and investors thin on relatively undifferentiated networks
  • EKs and Ethereum 2 will launch us into a low-cost innovative cycle for crypto
  • Same thing happened in the 90s with internet infrastructure, but once everything blew up, the cost to innovate was so low it led to the internet explosion.
  • “supperior tech” doesn’t necessarily justify a network valuation similar to Ethereum’s
  • Network effect of third-party distributors can outweigh superior tech
  • Investors in EKs hold most of the tokens and are itching to cash out. They will cause downward price pressure on the tokens
  • Miners will only sell coins below cost if they lose faith in the value of the tokens and network
  • The EKs that emerge at the end will be strong. Sometimes a fire has to burn the forrest to make room for saplings to grow.
How to think about value
  • Long-run, markets allocate value along the lines of costs. So to find value, look for costs. This insight is useful because observing costs is more practical than speculating on future profits.
  • Value capture does not equal high returns. If capital requirements are high to achieve value capture, returns will be low
  • Likewise, you can find high returns in markets with low value capture (TAM) if the capital allocation costs are low
  • To value crypto, must consider cost of capital to suppliers, cost of capital for investors, and value to users
  • Keeping things in balance is key to a healthy network (and long-term returns)
  • Protocols are higher cost than applications -> markets will award value to those who take on more risk and cost -> right now that’s the protocol layer
Ethereum and the seven dwarfs
  • In 1960s, the computer market was IBM and the seven dwarfs who all competed ruthlessly to gain a competitive advantage over IBM, but they all failed. Ethereum is the IMB of smart contracts. It might not be the best tech, but it works well enough and has a large enough distribution that it will be hard or impossible to unseat
  • EKs are focusing on superior tech, but Ethereum has (and likely always will have) distribution
  • Most people are still building apps on Ethereum knowing it’s not the best tech
  • It’s possible there are higher returns in niche markets (gaming, fashion, etc.) where you can grab a bigger piece of a smaller pie
Web vs. crypto service model
  • The more we decentralize cost structure, the more we distribute value. This typically has a linear relationship
  • Web services are centralized in companies like Google and Facebook and bear all the cost of productions
  • Crypto networks are kinda like global franchise agreements
  • The data in crypto is public and everyone has a copy, whereas Google/Facebook own their data. And personal info in the blockchain is encrypted in a way that only the user with a key can decrypt it
  • In current playbook, investors and entrepreneurs assume all risk in exchange for all the future equity value. In crypto, anyone can participate and for nearly any amount of value
  • For crypto, we need to optimize governance and networks to spread the risk and reward out as evenly as possible
Sovereign crypto networks
  • Sovereign states are countries that can run independent of outside influences. Sovereign crypto networks are the same concept.
  • Crypto networks need to be sovereign to ultimately fulfill what they’ve been promised to
  • The internet is somewhat sovereign. Local governments and ISPs can control local nodes, but it’s pretty much impossible to control all of them
  • The expansion of the internet broke up big telecom, but the nodes in the network are still “dumb” in that they don’t do anything except transfer data between A and B. There was a short gold rush in internet infrastructure, but once competition drove margins down, it enabled an application gold rush based off cheap data transfer and storage
  • In many ways, they think crypto will do to big web what the internet did to big telecom
  • Crypto communities are like real communities in that they thrive when in environments with lots of trust
  • Facebook’s libra network is an example of a non-sovereign crypto network. Since it’s financed by American firms, Congress has the power to control it. This is bad and will likely not work. Something like Bitcoin or Dai can’t be controlled by governments
  • Elements of crypto sovereignty:
    1. Decentralized supply:
      • Only online service with 100% uptime
      • Scales naturally as demand grows
      • Geographic diversity matter in case of government regulation
      • Some coins are considering geographic diversity of nodes in mining rewards (Filecoin)
    2. Decentralized demand:
      • Look at total number of users, where they are, and also the number of interfaces using the network
      • Allows for different apps to target different users across geographies, regulatory environments, languages, cultures, etc.
    3. Decentralized capital and market structure
      • This is an under-appreciated aspect of a network’s success. There needs to be exchanges, wallets, etc. that allow users to adopt the tech
      • Also important for token to trade against lots of currencies and in many geographies
      • Need investors in market to supply liquidity
    4. Decentralized governance
      • When projects are centralized in a company (like Kin coin), the SEC has the power to go after them. They can’t go after open-source developers and miners spread over the world.
      • In most cases “decentralized enough” will work and keep regulators at bay (Ethereum)
  • Largest and most successful networks will be those that are most sovereign, but at this point there is still a problem of too much centralization in a lot of networks
  • Decred is a Bitcoin fork that adds proof-of-stake to the consensus protocol and also dictates 10% of each block reward to a fund that goes towards core development. Bitcoin, on the other hand, is managed by a few central developers (not decentralized)
  • The more we decentralize power, knowledge, wealth, etc. across larger numbers of people, the healthier the network ecosystem will be in the long run
FOAM investment thesis
  • Google pretty much has a monopoly on the data layer of GPS and collects and sells user data without consent
  • FOAM:
    1. Reduces reliance on satellite services
    2. Provides open access to key meta data like geo-coding or points-of-interest
    3. Guarantees personal access through the user of open standards
  • FOAM will service all the new blockchain apps that right now have no way of integrating location services
  • You stake your FOAM coins to register points of interest on maps. The more you stake, the more the point persists as you zoom out
  • Can also be rewarded for verifying points of interest
  • Dynamic proof-of-location is being developed to track your location privately on the blockchain.
  • Seems like they could work well with Helium?
  • Rewards are based on how much points are used, so miners are incentivized to stake early to claim spots and to put enough coins down to make sure the point stays in veiw after an appropriate amount of zooming out
Cryptonetwork governance as capital
  • Governance of crypto networks should be viewed as capital assets
  • Capital offers control over distribution of economic resources across a group of people. Human capital controls service, political capital governs rules of markets, social capital drives human attention, etc.
  • Equities are very clear in that ownership of stock gives you rights to a share of profits and assets. For crypto governance, things will be murkier and more intangible
  • You need good cryptoeconomics (rules of the system) and governance (who gets to decide the rules, change them, etc.) for a healthy crypto network
  • Currencies like Decred are setup to do well on a long runway with solid governance models in place
  • Crypto networks are creating a new kind of capital—network capital. It’s cheap to produce and natively digital.
Crypto hedge funds vs. venture funds
  • A key decision for new crypto networks is deciding how to raise money
  • Hedge funds offer some liquidity. LPs can usually take their money out (or part of it) after some lockup period. And new people can enter the fund at any point. Always a risk of mass withdrawals
  • Venture funds use “committed capital,” which means they call some % of committed capital over usually a 10-year period. Capital is only redistributed when investments are liquidated, LPs can’t take their money out.
  • Both use “2 and 20” fee structure, but hedge funds base 2% management fee off quarterly value of assets (fluctuates with deposits and performance) while VCs base it off the total committed capital, which leads to a reliable source of funding. Also, 20% fee for hedge funds is calculated per quarter based on profits, but VCs only get paid out when they exit positions and return capital.
  • Hedge funds have greater incentive to maximize short-term profits, VCs longer-term profits.
  • Having both in the ecosystem is important to develop healthy economies–one provides short-term liquidity and price discovery while the other promotes innovation in the long-term
  • Hedge funds can also short investments, going against long-term stability of the network. VCs can’t do this by law
Funding crypto networks
  • Funding teams and funding networks are both important for crypto, but are very different
  • Funding teams takes place in series, much like traditional companies raise VC money
  • Teams can raise funds through locked-token sales, company stock, and other illiquid forms of capital
  • Funding networks is done through buying tokens over exchanges. See diagram in post. VCs tend to fund teams while hedge funds tend to fund networks
  • They try to provide some liquidity to networks they invest in when there’s selling pressure, and they dump off some assets when things get too heated. It can mean short-term profit loss, but they think in the long-term it will add to network stability and help adoption
The cryptoeconomic circle
  • Crypto networks are micro economies organized around a single service
  • Miners (demand side), users (supply side), and investors (capital side) come together to form a healthy network
  • Miners execute a consensus protocol, which allows users to use some type of service. Investors provide liquidity to miners, facilitate exchange, and help capitalize networks when they need help.
  • Three key relationships: miner-user, investor-miner, investor-user
  • Miner-user:
    • Miners are compensated for their work executing the consensus protocol in tokens. Users extract a service from miners paying in tokens, which creates demand and miner liquidity.
    • Supply side is distributed, which lowers costs and spreads it out
  • Investor-miner:
    • Short-term traders provide liquidity for miners so they can cover operational costs and grow.
    • Long-term holders capitalize the network and support price growth and adoption
    • Crypto network is fully capitalized when the cost to mine is breakeven with the current price.
    • In the beginning, investors help bootstrap demand so that supply side has liquidity and can scale
  • Investor-user:
    • Usually demand is driven by users, but with proof-of-stake, it’s also driven by supply side.
    • Higher prices support token holders and allow them to use the network service more
  • Isolating these roles and thinking through how they interact in each crypto network is essential to spotting healthy ready to grow.