There is a growing hope in cryptocurrency circles and beyond that Proof-of-stake (PoS) can contribute the benefits of cryptocurrency while avoiding the energy-consuming properties of Proof-of-work (PoW). This hope will eventually be dashed, as Paul Sztorc discussed this issue in 2015 in “nothing is cheaper than proof of work.” PoS simply obfuscates costs, but cannot eliminate They do. Here, I want to talk about why PoW is like democracy: it is the worst possible solution, but it is the only one that works.
This paper will not discuss whether the security of cryptographic currencies is “worth” the cost, but only whether PoS is an efficient approach.
The best cryptocurrencies will get security the cheapest way
To reasonably compare PoW and PoS we need to take into account both the costs (carbon footprint) and the benefits (blockchain security). So, our question can either be “To get a given security, is the environmental cost of a proof of stake lower” or “Given the cost, can a proof of stake get a higher security”? These two questions are logically equivalent. For the purpose of this paper, we will take the second line of questioning.
Axiom #1: The cryptographic currency that is most friendly to the environment should be the one that is most efficient when buying security.
Money is not magic. For money to circulate it must have a certain level of security, and the way to get that security varies in efficiency; and the most efficient technology will also have the smallest environmental carbon footprint. This is a very useful perspective to think about, because it reminds us of a profound economic law: marginal benefit equals marginal cost. A blockchain that spends $X on security means that miners will spend a total of $X to compete for that reward.
Axiom #2: If a cryptographic currency buys $X worth of security, its verifier (for the reward) will have spent a total of $X worth of money.
This is an unbreakable economic law about competition. Any miner who spends less than he gets will be beaten by those willing to spend more; and any miner who spends more than he gets will eventually go out of business. A blockchain that spends $X on security will always eventually lead to miners spending $X to defend that blockchain. The weight of the different types of expenses may vary, but all of them add up to $X. The difference is the same as a pound of cotton and a pound of iron.
Electricity is only part of it, not all of it
When people say “Bitcoin wastes energy”, they mean that electricity is needed to power the mining equipment that produces Bitcoin blocks; ironically, on the one hand, this is clearly not all of the energy used to run the Bitcoin network, but only a fraction of it; on the other hand, this is the fraction that is most directly used to protect the blockchain and is therefore the least wasteful. To fully measure Bitcoin’s carbon footprint, we also have to consider the cost of hardware (the cost of manufacturing and end-of-life) as well as the cost of normal business operations (having to renovate a decent office and fly in and out of Bitcoin conferences). These latter two types of costs are much more vague and difficult to estimate. But they are just as important to understand what the overall system will entail.
Axiom #3: To consider the carbon footprint of a system, you must consider all aspects of the verifier’s expenses.
There is waste in each of these three types of activities. The Bitcoin network will inevitably have orphan blocks, and then the energy to mine that orphan block is wasted. Some hardware is built, but it doesn’t operate as efficiently as it could. Some operational overhead is necessary, and some is just a change of pace for a cup of tea. The more closely an expense is tied to the actual work of verifying the blockchain, the more efficient it is in exchange for blockchain security. Similarly a machine is more efficient when it has fewer degrees of freedom: one more part is more useless work. The same holds true for economic systems – the more structure, the less efficient.
As an example: traveling to play nice with some big customer can be abstracted to improve the efficiency of the whole business, and therefore help to buy security — but it does so very indirectly, and its buying security may not be as effective as the equivalent expenditure on electricity. The more a cryptographic currency spends its security budget directly on electricity, the more secure it is; the more it spends on commercial operators, the less secure it is. The CEO of a mining business may be necessary for that business itself, but they contribute little to the security of the network.
Axiom #4: The more an expense is associated with the actual action of verifying a blockchain, the more efficient it is at buying security.
So, sometimes you see people discussing ASIC 1 and thinking they corrupt the environmental impact of a cryptographic currency, but in reality it’s just the opposite. The specialization of mining equipment allows more arithmetic to be generated per unit of hardware, which means blockchain security budgets are spent less on hardware and more on power. This is good for efficiency — ASIC mining is fundamentally cleaner than traditional CPU mining because ASICs are more efficient (i.e., cheaper) at converting energy into security.
This is why people misunderstand the environmental impact of “Proof-of-space” type cryptographic currencies like Chia (which use memory rather than computational power to authenticate the blockchain). Using hard drives as a scarce resource for generating security does not reduce costs, but simply allows the blockchain’s security budget to be allocated and spent on hardware from electricity. If you look at power consumption alone to measure the carbon footprint of cryptographic currencies, you’ll feel as if we’ve made huge strides — but if you factor in the expenditure on hardware, you’ll see it’s a big loss of efficiency.
There are a lot of unused hard drives in the world today, so Chia is getting some subsidy. But don’t expect this to last forever. By the time some sort of cryptographic currency succeeds in monetizing computer memory, what has already happened with GPUs will be repeated in the related hardware space. More importantly, eventually this system will be worse than a system of equal size but based on PoW, because the security budget is spent on developing the hardware, and that is less efficient than a security budget spent directly on power.
- ” Tom’s Hardware has already reported on the skyrocketing prices of Chia and the shortage of high-capacity mechanical drives. But what may be more critical is the SSD (shortage), because you also need SSDs to make Bram Cohen BinggoTM cards that can cram large drives” –
Conclusion #1: The most environmentally friendly cryptographic currency is the one that spends its security budget on the matters most closely related to authentication.
Locked capital is also a real expense with real overhead
Chia and other blockchains that use spatial proofs are still relatively easy to be able to compare to PoW because you can easily imagine old hard drives stuffing landfills, so you can easily be convinced that they “really” have an environmental impact. But lost capital also has a real impact on the environment. Because that money could be spent on carbon capture technology and carbon efficiency research, and so on. People “feel” they are free (not consumed) simply because humans are not good at thinking about time-value. If capital was really free, then miners would get it without paying anything.
Sometimes you also see people asserting that proof of entitlement is also wasteful from an individual perspective, but is a savings from a societal perspective, as Dan Robinson says.
- “Suppose your religion requires you to prove your sincerity by your own sacrifice. You could throw a cow into a volcanic crater, or you could gift it to a stranger. Both things are equally costly to you (and therefore both satisfy the commandment), but the social cost of the first is higher.” —
The truth is just the opposite. Locking up capital is not a waste for you personally, because the PoS pledge algorithm compensates you for it – that capital is not thrown into a volcano, it earns you a return at the expense of other ETH holders. Locking up capital is completely different than giving it away to strangers, because there are no strangers taking over the funds at all. The validator doesn’t “distribute their capital equally to everyone”, in fact, it’s everyone who pays equally for the validator’s work (in the form of newly mined ETH). And, thanks to Axiom #2, we know that the payout received by the verifier is exactly equal to the cost of those funds.
On the other hand, locking up capital is a waste from a social perspective, because it can no longer be used to build factories, fund research, and do everything that is good for society. Consider the financial crisis of 2008 – no factories were destroyed, no houses collapsed, no physical assets were lost. The consequence was simply the loss of capital – but that is still a very, very heavy social cost. Capital is a special form of information accumulation, coalescing information about the optimal allocation of resources. The loss of information is not the same as the loss of something tangible, but it is a loss all the same, if not worse.
So, keeping in mind Axiom #1 and Axiom #2, a fair comparison of PoS and PoW would look like this.
Proof of entitlement does not (and cannot) eliminate miners’ expenses, it just shifts the portion of PoW system expenditures on electricity to capital expenditures. Locking in the externality of capital, as opposed to how the externality of power consumption compares, is complex and subtle — but the bad news is that proponents of PoS systems tend to pretend that power is the only cost (of running the system).
- The intentionally misleading comparison, taken from the Ether Foundation blog. –
Disappointing. The Ether Foundation is well aware that “average energy consumption per transaction” is a completely meaningless metric, and using this data in this way is intentionally deceptive. The point is not at all whether PoS systems use less energy – they do, but that’s not the whole story. The meaningful discussion is: is it better to consume capital, or is it better to consume electricity? Any theory that supports PoS without discussing the cost of capital is fundamentally wrong.
Efficiency Comparison: Pledging vs. Mining
In the PoW/PoS debate, PoS proponents often have an advantage: there has never been a fully decentralized proof-of-stake system. This also means that PoS proponents are free to fantasize about how efficient and elegant the final solution could be. None of the known mining algorithms are perfectly efficient (e.g., there is a “selfish mining” problem), and again, there is no reason to believe that PoS systems will be infallible – but until the solution is finally fully formed, PoS proponents do not have to consider the specific tactics that validators might use to trick the system until the solution is finally fully formed.
The more opportunities there are to trick the system, the looser the relationship between the security budget and the actual verification. Because “no one has ever tried a real proof of interest,” we don’t know how well the security budget can be translated into locked-in capital in the first place, much less how efficient the locked-in capital is at generating security. By definition, PoW is tightly coupled to verification, but PoS, as of writing, has not been fully defined.
Observation #1: PoW is tightly coupled to security. But it is not clear how far PoS can go.
Externalities: Pledging vs. mining
Assuming we all agree that locking capital and mining are equally efficient from a security efficiency perspective, that leaves only one issue, “externality”. This is where I think rational discussants can still disagree about the trade-offs involved.
In the short term, the PoW system is going to compete with other projects for energy use, but in the long term it will incentivize the development and production of cheap energy, and that will gradually lower the cost of electricity, bringing benefits to everyone (both for financial purposes and environmental protection). Cheaper electricity will also lead to new technologies and more production.
Equity justifies the use of capital in the short term in competition with other projects, but it does not provide an incentive to create more capital, so it will gradually raise the cost of capital. A higher cost of capital means fewer projects — fewer factories, and less research.
Whether you prefer the externalities of PoW or PoS may depend on whether you think it is better to solve the environmental crisis by investing in technological development, or by saving on current consumption.2 To save the planet, should we make cheaper renewable energy, or should we make everything more expensive?
Observation #2: Proof of workload encourages future investment. Proof of Stake, on the other hand, curbs exploration.
PoS doesn’t even work.
As we said earlier, we don’t yet know if a fully decentralized proof-of-stake system can be implemented. Existing cryptocurrencies that advertise themselves as PoS systems basically rely on some centralized coordinator or “checkpoint” (checkpoint). Since Ether has been launched from 2015, they have been trying to convert to a fully decentralized PoS system, and up to now, they have only launched Phase 0, which just supports verifier pledges …… Once you lock your money in, you can’t go out or spend it.
There is a lot of debate about whether PoS systems can be secure enough (see “long-range attacks”, “nothing-at-stake”, “workload across time/equity is the opposite “). These are open questions for research, not a to-do list for the implementation process. We don’t know how long it will take to figure out a good solution, and it is doubtful that there is even a good solution. Check out this 2017 post at.
PoS may “work”, it just works in a pathological way. poW would favor those who can get power cheaply and reward them with capital. Not exactly equal, but not stuck in a positive feedback loop. PoS, on the other hand, favors those who can get capital cheaply and rewards them with capital, which creates a loop. Those who have money will become richer, and the richer they are, the easier it is to become richer.
If proof of entitlement “can be started” but the end result is that all power is concentrated in the hands of a small group of super-rich institutions and they can control the system in perpetuity, then it doesn’t solve the problem that cryptographic money is trying to solve. Then we might as well continue with the current system of central banks and “insiders who benefit from the Cantillon effect”. We want cryptocurrency to break the system, not to be old wine in a new bottle.
Observation #3: Proof of workload already exists and has proven itself to continue to grow. But proof of stake is still an unsolved problem.
Axiom #1: The cryptographic currency that is most friendly to the environment should be the one that is most efficient when buying security.
Axiom #2: If a cryptographic currency buys $X worth of security, its verifier (for a reward) will have a total of $X worth of spending.
Axiom #3: To consider the carbon footprint of a system, you must consider all aspects of the verifier’s spending.
Axiom #4: The more an expense is related to the actual action of verifying a blockchain, the more efficient it is to buy security.
Conclusion #1: The most environmentally friendly cryptographic currency is the one that spends its security budget on the matters most closely related to the verification effort.
Observation #1: PoW is tightly coupled to security. But it’s not clear how far PoS can go.
Observation #2: Proof of work encourages future investment. And proof of equity curtails exploration.
Observation #3: Proof of workload already exists and has proven itself to continue to grow. But proof of equity remains an unresolved issue.
Unfortunately, after a detailed examination, we learned that most of the “benefits” of PoS are only seemingly glamorous, and that there are many significant open challenges. It remains, at least for now, a utopian fantasy rather than a realistic option. PoW, on the other hand, despite its drawbacks, is still the best known solution (indeed, the only solution) for forming a network that does not require trust in a centralized authority.
Specialized integrated circuits: computers that are customized but can only do a limited type of computation and nothing else (ASICs developed for mining specific algorithms are therefore very efficient at mining).
If we look realistically at human nature, the only way to reduce consumption on a global scale is to improve technology or reduce population. Any philosophy that advocates a direct “reduction in consumption” is either very naive or very scary.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/beyond-power-pos-is-not-a-savior/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.