The Energy Debate: PoW vs PoS
In this post, I'm returning to share my thoughts from Matt Levine's "The Crypto Story." I've picked out a few quotes that stood out to me because they tersely underline points and arguments worth understanding in greater detail.
Here are three quotes which I took from the "What Does It Mean?" part, where Matt introduced readers to Ethereum, the second largest cryptocurrency, and its new consensus protocol — the proof-of-stake (PoS).
[PoS] is more efficient, in two ways. First, it uses less electricity. Burning lots of electricity to do trillions of pointless math calculations a second, in a warming world, seems dumb. Proof of stake uses, to a first approximation, no electricity.
[The] second [way]: PoS more directly measures your stake in the system. You demonstrate your stake in Ethereum by 1) owning Ether and 2) putting it at risk to validate transactions. To take control of the PoS system and abuse it for your own nefarious purposes, you need to own a lot of Ether, and the more you own, the less nefarious you'll want to be.
You can make it even easier on yourself. Instead of downloading the software to run a full Ethereum validator node, and depositing 32 Ether, you can hand your Ether over to someone else and let them be a validator.
Let me give some context to explain this better. A consensus protocol is a fancy name for a method to verify transactions in a blockchain network. More specifically, it is a set of rules everyone in the network agrees to follow. Bitcoin uses a proof-of-work (PoW) protocol that requires electricity to solve mathematical problems to validate transactions (here is an article that explains the PoW concept nicely, or you can check one of my earliest articles).
Meanwhile, Ethereum switched from PoW to a PoS consensus mechanism. The PoS protocol is a bit more complicated; holders of Ethereum coins (ETH for simplicity) volunteer them by depositing coins into special applications to have a chance of becoming validators. The protocol randomly chooses a validator at fixed intervals to create a block with transactions and sends it to a network of validators for review. If the network agrees on the legitimacy and validity of the block, it joins the blockchain, and the original validator receives a reward. The critical difference between the PoW and PoS protocols is that rather than spending and "wasting" electricity to validate transactions and secure the network, users will lock coins in the system, putting them at risk. The risk will prevent users from acting wickedly; hence, the blockchain will be safe and sound.
[PoS] is more efficient, in two ways. First, it uses less electricity. Burning lots of electricity to do trillions of pointless math calculations a second, in a warming world, seems dumb. Proof of stake uses, to a first approximation, no electricity.
Wasting electricity on "trillions of pointless math calculations" is the most common critique of Bitcoin and mining. I've already discussed Bitcoin's electricity consumption in two previous articles (you can check them here and here), yet today, I'm going to dig deeper into this debate.
Solving mathematical problems — otherwise known as mining — is needed to secure the network and prevent it from nefarious actors. Miners consume electricity to produce and add blocks with transactions to the blockchain, and in return, they receive rewards — bitcoin and users’ transaction fees. Suppose a miner tries to trick the network and include fraudulent transactions. In that case, the network will not accept or validate the block, leaving the miner empty-handed and with a hefty electricity bill. In other words, the PoW protocol gives miners monetary incentives to play by the rules if they want to remain in business. Therefore, electricity ensures that only valid and legitimate transactions are added to the blockchain, thus creating a truthful set of records, as Saifedean Ammous pointed out in Chapter 10 of "The Bitcoin Standard".
If we look at the PoS protocol, the security mechanism resembles that of the PoW to some extent. Validators risk their staked coins when they propose a new block with transactions — if the block doesn't meet the rules, a validator will lose his coins.
However, what makes PoW superior to PoS is the energy itself. I know it's a bold claim but let me defend my statement.
Large quantities of electricity and processing power are necessary to impose real-world physical prohibitive costs on people solving mathematical problems, as Jason Lowery described in the "Bitcoin Fundamentals Podcast". It means that people must use physical resources from the real world to create new blocks and mine bitcoin in the digital world. You can think of this as a bridge between two realms. In contrast, the PoS protocol uses digital tokens that do not exist in our world to secure the network that issues these tokens.
Energy is fundamental to our universe, yet the conservation of energy makes the universe rational and makes it work, Michael Saylor, chairman and ex-CEO of MicroStrategy, said in the subsequent episode of the "Bitcoin Fundamentals Podcast":
Money without energy is credit. A commodity without energy is coupons. In cyber-space, an object without energy is an image.
(I highly recommend listening to both podcast episodes as they are part of a series that Preston Pysh, the podcast host, made to discover why PoW and Bitcoin's energy use are crucial for changing the world.)
When an object requires energy to produce, it becomes costly to copy. The costs limit the supply of the object, making it scarce and giving it credibility (a physical apple you can eat and not a coupon to get the apple, which is not edible).
If I give you a coupon promising an apple next week, and you use your phone to send the coupon to a hundred friends, chaos and fighting will ensue when everyone shows up to get one apple from me.
The same is true when we look at money: chaos is inflation and its consequences, using that phone is money printing, and the apple coupon is money.
The only way to avoid chaos is to inject energy into money to make it both scarce and credible. For those thinking connecting money to energy is something novel and out of a fantasy book, I can reassure you that it's not. In "A Bullish Case For Bitcoin: My Take on Bitcoin-Dollars," [LINK J] I pointed out:
Energy standards have been common throughout history, for instance, the gold standard. Gold is valuable because it is hard to find and requires energy to extract from the ground. And when money was connected to gold, one could only produce more money by putting more energy to extract more gold.
Let me draw your attention to "hard to find." Thanks to the real-world physical prohibitive cost, money units whose growth rate can't be easily meddled with are known as hard money. The harder the money, the better it is suited to perform monetary roles (store of value, a unit of account and medium of exchange), as Saifedean Ammous states in "The Bitcoin Standard".
However, Michael Saylor points out that money injected with energy is credible, yet not perfect. To perfect money, the monetary system needs a decentralised ledger "of Bitcoin type system" that ensures the total money supply is capped at 21 million BTC, and no one else can produce more. Just like I mentioned before, money is a coupon. Hence digital money is a digital coupon. Michael Saylor argues that the invention of the Bitcoin protocol "that the energy circulates within" moved us from a simple digital coupon to a digital scarcity, creating the hardest money currently available.
So, to quickly summarise, energy is essential in order for Bitcoin to keep its monetary properties. When we remove energy from a crypto network, we remove the real-world costs to create new coins or copy the network itself. As a result, there will be no scarcity and no value (Michael Saylor called it the economic hazard). In addition, when we destroy the bridge between the real and digital worlds, it is necessary to simulate everything happening in the real world - physical forces, human behaviour, the universe etc. Such simulation is highly complicated, extremely expensive and sensitive to security threats. It takes only one mistake to bring the world down.
Building a simulated world and keeping it secure will only be possible when a single organised software development company takes charge — raises the initial capital and coordinates the work of programmers. Doesn't this sound like an ordinary tech company? According to Michael Saylor, such arrangements will eventually lead to power concentration in the hands of software developers, which turns both the company and the network tokens into securities.
If they have the ability to slash or to grant permission or deny access. Or, for example, to let you withdraw your tokens. At that point, the software company has become so central to the network that you've created a security. It becomes an investment contract per the definition of securities law because you are relying on the efforts of others. And it's pretty much impossible not to rely on the efforts of others when you require software engineers to make the sun shine and make the planets revolve around the sun, and enforce gravity.
The second quote is as follows:
[The] second [way], PoS more directly measures your stake in the system. You demonstrate your stake in Ethereum by 1) owning Ether and 2) putting it at risk to validate transactions.
Ethereum developers are yet to figure out a way to allow users to withdraw coins from staking applications. So, the staked ETH looks like shares in a company that gives you voting rights and can't perform monetary functions.
Besides, when you have to rely on the efforts of others to maintain the simulated world to make the protocol work or that they will add features to make the protocol work as intended requires a lot of trust. Will it run smoothly from now on? Do I trust developers to put the interests of the community and the network first?
These questions defeat the primary idea behind Satoshi's protocol. Satoshi wrote in the conclusion of the Bitcoin whitepaper:
We have proposed a system for electronic transactions without relying on trust.
Moving away from the PoW to PoS protocol feels like one step forward, two steps back and I will explain why. As Jason Lowery explained in the podcast interview, to secure software, the developers create an abstract power hierarchy where certain positions have certain ranks, e.g. admin rights. Since everything in the software is abstract, we use an abstract power hierarchy to determine who controls resources and how to legitimate ownership within the software. Software hacks are just problems with people systemically abusing the abstract power hierarchy. The main challenge with solving these problems is that it's impossible to impose real-world physical prohibitive costs in cyberspace. It was a problem until Adam Back invented the hash function, one of the underlying technologies of the PoW protocol.
The hashing function creates a physical power hierarchy in the real world that is more inclusive and decentralised than the abstract power hierarchy. More importantly, the former is more secure as it can't be as easily exploited. Jason and Preston go deep into the discussion between these two power hierarchies at the start of the podcast. To stay on course, let's bring Bitcoin and Ethereum.
If a dishonest miner wants to take control of the Bitcoin network, they must acquire at least 51% of the total processing power. However, honest miners can quickly stop this entity by expanding their processing power and keeping the intruder at below 50% level. Happy days!
The same approach is not possible in the PoS protocol. The processing power "comes" from staked coins. Let's assume a fixed total supply of coins and users stake every single coin. If one entity gains control of 60% of the supply, it can split it across various validations and use them to withhold specific transactions. The honest validators that control the remaining 40% of coins will be unable to increase their stake to push the evil entity out. Therefore, everyone must trust each other in such a network, and trust is inherently insecure.
This brings me to the third quote about how it's easy to participate in transactions validation:
You can make it even easier on yourself. Instead of downloading the software to run a full Ethereum validator node and depositing 32 Ether, you can hand your Ether over to someone else and let them be a validator.
It is easier to become a validator on the Ethereum network than buying specific machines for PoW and setting them up in your basement to mine bitcoin. If you don't have 32 ETH — around $38,000 — you can stake your Ethereum through a staking pool. Sounds excellent, doesn’t it? More people have a say in the creation of new blocks, which should make the network more decentralised. However, the reverse is true. Since Ethereum transitioned to the PoW protocol, greater centralisation has occurred in the network. As you can see on the chart below, two staking pool companies — Lido and Coinbase — hold 44% of the staked Ether (the top 4 companies have a 59.3% share).
If these companies decide to collude and push their own "agenda”, the remaining 40.7% of validators will be powerless.
At the moment, the Ethereum users must trust the staking pool companies and the Ethereum foundation to act in the best interest of the network and community. And this year we've all seen what could happen with centralised companies that lose sight of prudent management and take excessive risk.
I can't end the post without saying a few words about the electricity type Bitcoin uses and the emissions from its mining operations. In 2021, Bitcoin consumed 0.05% of the global energy and contributed 0.08% of the global carbon emissions, according to the CoinShare report. Although these numbers may look huge, the Bitcoin community is working tirelessly in helping the energy transition and the reduction of greenhouse gas emissions. In the US, Great American Mining developed trailer-mounted miners that can be easily set up in oil fields. These mining rigs are powered by dry natural gas that otherwise flares or escapes into the atmosphere. Meanwhile, bitcoin miners in Texas are working closely with an electricity grid operator to improve reliability and efficiency of the grid. Canadian MintGreen is working on solutions to repurpose heat from bitcoin mining for heating homes.
Energy is an integral part of Bitcoin, making it the world's most scarce and secure asset. Not only has Bitcoin brought monetary and technological revolutions, but it is also helping to improve electricity system resilience as the world leaps towards renewable energy sources.
If you’ve been enjoying my writing, I want to encourage you to subscribe to the notification list so that you never miss when I post a new blog!
It may even prompt me to write more often :) And now to the new post!