(Any views expressed in the below are the personal views of the author and should not form the basis for making investment decisions, nor be construed as a recommendation or advice to engage in investment transactions.)
In September, Ethereum, the 2nd largest blockchain protocol completed what has been dubbed “The Merge”. It transitioned from a Proof of Work (“POW”) consensus mechanism to a Proof of Stake (“POS”) consensus mechanism. It is an impressive technical accomplishment, which can be best understood as a software upgrade. However, what is lost in the hype surrounding the merge is the significance of the problem the merge is attempting to solve. The merge, or POS, is seeking to achieve distributed consensus in a permission-less, censorship resistant manner. That is a mouthful. Said differently, all public blockchains are effectively trying to create a digital ledger that accepts and settles transactions from anyone at anytime. To achieve this the ledger must be replicated and validated by all network participants to ensure a consistent view of token ownership. Further, the blockchain must have a mechanism to determine the time ordering of transactions. Yet, this is significantly harder than it sounds due to something called the “double spend problem”. The double spend problem is the challenge of individuals spending the same digital token multiple times.
In blockchains time-ordering is important to prevent a recipient from spending a token simultaneously with two different people. In traditional ledger systems, such as a bank account, we rely on centralized entities to establish consensus, or said differently, to ensure each transaction is valid and that the unit of account is only spent once. Here, I detail some of the trade-offs of this model. In short, it can lead to rent seeking behavior, as the act of verifying the ledger is costless for the central intermediary. Further, the current model centralizes control, giving power to intermediaries to censor or alter the rules (this is an important point and will become more apparent later in this piece). However, it is important to note that traditional / ordinary consensus is much easier and far more efficient. As such, one must believe that the trade-off or what is being secured via distributed consensus is “worth it”. The fundamental question for all blockchains is, is the asset the blockchain secures worth a more costly and inefficient, relative to traditional methods, yet potentially significantly more secure and censorship resistant settlement mechanism
As the proliferation of public blockchains increases, several consensus mechanisms have emerged. However, the two most popular approaches are i) Proof of Work and ii) Proof of Stake. In what follows, I will provide a crash course on both approaches and the trade off each makes to solve the same problem.
“Elliott, this is a 2K word piece, what is the TL;DR?”
The Tl;dr is that there is no “right” consensus mechanism; instead, it is context dependent. In the case of Ethereum, it’s stated goal is to become a world computer that facilitates a diverse ecosystem of decentralized applications. In this context, Proof of Stake might make sense. Bitcoin, the largest public blockchain network, uses a Proof of Work consensus mechanism. While unlike Ethereum, there is no leader to make proclamations about the goals of the blockchain, the narrative that has driven Bitcoin adoption over the last 13 years is that bitcoin is a digital bearer asset, or digital property akin to gold. The properties of bitcoin are supported by its underlying computer code, which most importantly has a fixed cap on supply. Bitcoins’ fixed supply makes it finitely scarce, a property that has not existed in any other commodity in the history of humankind. Yes, it sounds a bit dramatic, I know, however, it is true. In the context of a creating a scarce form of digital property a Proof of Work consensus mechanism is more logical, as there is a real-world cost to publish new transactions to the public ledger, which incentivizes all network participants to “play by the rules” (i.e, not submit double spend transactions) and thus supports the properties of the underlying code
What is Proof-of-Work?
Put simply, Proof-of-Work consensus entails the authentication of transactions by a collection of transaction signers called miners. Miners do not identify themselves to other participants and may enter or leave the network at anytime and at no cost. Miners produce signatures by a process called mining wherein they produce proofs of work on successive blocks of transactions. Before going into the details of how mining works, there are a few characteristics that are important to highlight:
1. Miners are anonymous. The inability to identify who the individual miners are makes it incredibly difficult to bribe them to produce invalid transactions to the ledger or influence them to censor / not allow certain individuals to participate in the network. A counter to this point is that mining facilities are easily geographically located and thus could be targeted for attack by a bad actor. However, this risk is mitigated by the diverse geographic presence of miners. It would be terribly difficult to attack the entire network at one time. In the event the entire network was attacked, there is nothing that prevents new participants from entering the network, which is the likely outcome, as the economic incentives to mine would increase
2. Miners can enter or exit the system at any time at no cost. Network participation is permission-less
Again, the problem that POW is trying to solve is the ability for a decentralized network to reach agreement on the time ordering of transactions. Time ordering is important in real-time, but most important is the ability to serve as a historical record of transactions overtime. Users who are new to the network or have been offline need access to historical transaction data. Users need to be assured that the transactions they are receiving occurred before any conflicting transactions
The Proof of Work solution to reaching consensus is to have miners expend energy in competition with other miners to append the next block of transactions to the blockchain. Miners use compute energy to solve a cryptographic puzzle, with the fastest solution winning the right to append the next transaction. Miners are incentivized to appended new transactions via a block reward of bitcoin and transaction fees for every block that they appended first. The speed of mining hardware (purpose-built computers) and access to low-cost energy are ultimately the key determinants of the success of miners. The cost of winning a block reward for a miner is simply the cost of equipment and the cost of energy consumed to “win” the right to post the next transaction
How are winners determined?
Network participants consider all series of historical valid blocks, in conjunction with new blocks that have been published by miners (collectively, the blockchain) and select the blockchain with the greatest amount of aggregate compute power. Put simply, the miner that expended the greatest amount of energy to solve the cryptographic puzzle wins
Why does this matter?
Energy is scarce and has real world cost (as we are seeing real time due to the war in Ukraine). The fact that miners must consume real world resources to append new transactions to the blockchain makes it costly for a bad actor to try and post false or double spend transactions. The economic cost of posting a new block introduces market incentives and game theory, which ultimately makes it more profitable and in the best interest of miners to only consume resources to post valid transactions. The only way to make a profit as a miner is by posting valid transactions and earning the block reward. It is a waste of energy and money to attempt to post invalid transactions to the blockchain, as majority of market participants are following the economic incentives of the rules and thus will reject invalid blocks - or blocks that include transactions that conflict with the history of the blockchain that the rest of the network is maintaining. In short, miners have constant skin in the game via the demand to consume real world resources to participate in the ecosystem. This positive feedback loop maintains the integrity of the Bitcoin protocol. In addition, it makes it very difficult for an adversary player to “attack” the blockchain as it would cost a significant amount of capital and energy consumption
What is Proof of Stake?
A proof of stake is a cryptographic proof of ownership. In the case of Ethereum, validators, which play a similar role to miners in Bitcoin, lock up, or stake, 32 ETH (~$40K USD, at current market prices), the native token to the Ethereum blockchain. Staked ETH affords validators the opportunity to append or validate new transactions to the Ethereum blockchain. In POS the ability to append new transactions is the result of random selection among only other individuals that have staked ETH, with your chance of being selected being proportional to the amount of ETH, as a % of the total, that you have staked. The winning validator is compensated via transaction fees
In contrast to Proof of Work, Proof of Stake is the idea that cryptographic proof, rather than physical proof can demonstrate that computation to validate transactions is costly. In Proof of Stake validators are incentivized to only propagate valid transactions because a) they are randomly chosen and therefore collusion is difficult, and b) their stake in the system incentives them to only sign valid transactions, as undermining the rules of the system would impair the staked capital
There are several key differences between POW and POS:
1. It is not costless to enter the system – you must have 32ETH to be a validator, which naturally places a limit on the ultimate decentralization of validators relative to POW. In fact, several centralized services have emerged that allow individuals that do not have 32ETH to pool their resources with others and the service provider will stake on behalf of clients. The pooling of staked capital has created significant centralization of validators
2. It is not costless to leave the system – currently your staked capital is locked in the protocol for a defined period (there is currently no way to withdraw staked ETH now – as the developers intend to build that functionality at a “later date”)
3. While the incentives in theory should preclude validators to play by the rules, it is costless to attack the network and attempt to append invalid transactions. The costless nature is since validators only must stake capital once. Therefore, in theory you can attempt to attack the chain at will for free. While there are mechanisms in place to disincentivize this behavior, the problem is that if the transaction history is easy to change you are inherently forced to trust that the history is what the validators say it is, which makes the centralization noted above an even more acute risk. This differs from POW where you do not have to trust anyone, and instead the correct transaction history is the one that has the most demonstrated work, or the max real world cost expenditure
My “hot take”
Admittedly, I don’t have that hot of a take on this topic. It is incredibly complex and far to early to appreciate the implication of either consensus mechanism. However, I do think POW and POS are fundamentally solving different problems and thus trying to compare them is likely a category error. I do believe that for digital scarcity to exist there needs to be a real-world cost to create new bitcoin. The production of scarce goods should not and cannot be free. If the production of a scarce good is free it will no longer be scarce over time as the producers of the good will always produce more if it is in their best interest, which it usually is (look up the Cantillion Effect). Ethereum is not attempting to be a scarce digital commodity. Therefore, perhaps, POS is a good solution. However, I do have concerns with the centralizing tendency of POS due to barrier of having to stake 32ETH and the compounding nature of being a validator (i.e., the transaction fees compound the initial investment, thus once you own a stake in the network it grows at a rate that makes it harder for new joiners to own as much stake in the network, no even accounting for the fact that the appreciation of ETH overtime acts as a natural barrier). Centralization is not ideal as it increases the risk of censorship and state capture, which is at odds with the point of decentralized public blockchains
Who did what this month?
Blackrock, the largest asset manager in the world with over $9 trillion in assets, is giving clients using its Aladdin portfolio management software the ability to trade crypto. The program will only support bitcoin to start. Link. Link. Blog Post.
Forty of the world's largest companies invested about $6 billion in blockchain companies between September 2021 and June 2022, according to a study by Blockdata that looked at the investment activity of the biggest 100 public companies by market cap. Samsung was the most active, investing in 13 companies. Google parent company Alphabet (GOOGL) participated in four funding rounds that raised $1.5 billion for blockchain and crypto-related companies, the largest value. Link.
Retail giant Walmart is allowing consumers to experience the metaverse through the launch of two new experiences in gaming platform Roblox. The move comes after Walmart filed seven trademarks at the end of December that signaled its plans to make and sell virtual goods in the metaverse. Link.
The Reserve Bank of Australia is expecting to complete its central bank digital currency pilot by mid-2023, according to a white paper published in September. Link.
Nasdaq is launching an institutional crypto custody service, pending regulatory approval. Link.
Autograph, the NFT platform co-founded by Tom Brady, is launching golf digital collectibles in partnership with the PGA Tour. The platform is expected to go live in 2023 and will sell NFTs of notable moments in golf history. Link.
Colorado is the first state to accept cryptocurrency for state taxes. Taxpayers can pay their taxes using PayPal, which supports Bitcoin, Ethereum, Bitcoin Cash and Litecoin. Link.
Crypto firm Chain signed a multi-year sponsorship deal with the New England Patriots. Link.