Wednesday, January 5, 2022

A tax on proof-of-work

The world is overpurchasing proof-of-work (POW) blockchains. How do we fix this? 

Let me quickly outline the argument for why the world is buying too much POW. Blockchains such as Bitcoin, Dogecoin, and Ethereum provide coin buyers with a special sort of security proof of work. POW requires huge amounts of electricity, so much so that Bitcoin and Ethereum together currently use up more energy than Italy.

It's not the energy-intensity of POW that's problematic. The issue is that the biggest buyers of POW coins speculators and gamblers care very little about POW security. What they value is the thrilling price movements that blockchain coins provide.*

Coin gamblers also have the option of buying non-POW blockchains. Non-POW coins offer gamblers the same wild price movements as Bitcoin, Dogecoin, and Ethereum. However, the security that these non-POW coins rely on requires far less electricity. 

On net, the world would be better off if all blockchain gamblers migrated away from POW coins and onto cheaper non-POW coins. The gamblers themselves would not be any worse off. They'd still get all the crazy up-and-down fun & entertainment as before. But the rest of us would benefit, since far less of the world's energy would be burned. That's what I meant at the outset when I said that we have a POW overpurchasing problem. Coin speculators are unwittingly over-gambling on energy-heavy POW blockchains and under-gambling on energy-lite non-POW chains.

Left to their own devices, it is unlikely that coin gamblers will migrate away from POW coins like Dogecoin and Bitcoin towards cheaper coins. (See here for why). Might it be worthwhile to adopt a policy that nudges speculators away from POW and into cheaper non-POW blockchains? 

A targeted tax on POW coins is one option. But how would we design this tax? I suppose we could tax people's holdings of POW coins while exempting their holdings of non-POW coins. Or we could tax POW coin purchases & sales on exchanges like Coinbase. Or would we could tax the POW miners.

The economics of taxation is not my strong point. So I'm going to farm this out the comments section: if we want to shift speculators away from POW blockchains, how should we design a POW tax?


*There is also a community of hobbyists and technologically-informed individuals who do indeed consume POW. That is, they can put out a coherent argument for what POW is and why they prefer it over other types of blockchain security. Unlike the gambling class, they are less interested in coin price fluctuations. But the size of the hobbyist community is dwarfed by the gambling class.

39 comments:

  1. Is there a % renewable energy of the network that you would rip up this idea?

    Like say bitcoin gets to the point it runs on 80%, 90% renewables?

    ReplyDelete
    Replies
    1. No, that wouldn't change my argument.

      Assume a POW coin that runs on 100% renewables and a non-POW coin that runs on 100% renewables. The POW coin uses up far more renewable energy than the non-POW coin. Gamblers are the main users of coins. They want something that goes up-and-down. They have no preferences for POW security over non-POW security.

      Society is better off with the tax because the coin gamblers are shifted onto a non-POW coin that provides all of the same gyrations, but on net less renewable energy is being burnt.

      Delete
    2. Got it that's fair. Not a bad idea. Maybe a tax on futures/leverage? That's where most the BTC gambling is I feel like. I know billions of $ in BTC is not gambling, it's for savings/transacting, lightning, etc. Seems to be that most gamblers are already gambling on mostly POS coins. Idk, just rambling.

      Delete
    3. It is arguable that Society is better off consuming less renewable energy. Consuming more renewable energy could make it more efficient (economies of scale)

      Delete
    4. "It is arguable that Society is better off consuming less renewable energy. Consuming more renewable energy could make it more efficient (economies of scale)"

      Switching to renewable energy is important. But don't forget the opportunity cost that comes with POW coins burning up renewables. Other industries would also like to use renewables. They can't because POW is using them. If we can shift speculators onto non-POW coins (at no cost to the speculators), then POW will use less renewables. That frees up those renewables to be used by other industries.

      Delete
    5. That would bid up the price of renewables in the short term, which would lead to more investment and consequently more renewable supply in the long term.

      IMO, the solution for this kind of problems is to find the way that the entity that generates an external negativity fully pays for it. That would be the responsibility of energy producers, not on the consumers (crypto speculators, christmas lights, home appliances on stand-by, etc)

      Delete
    6. "That would bid up the price of renewables"

      All sorts of industries are always competing to bid up the price of renewables. But we don't want industry x to do this at the expense of other industries y and z that are also bidding up the price of renewables, when industry x's primary output (price volatility) can be produced at a fraction of the energy cost. It's far more efficient to let the output that needs renewables to bid up the price of those renewables.



      Delete
  2. Great idea.

    First the government should buy put BTC options.

    Then it should dump all of the seized BTC and other PoW coins. Then it should enact this tax and use all the profits to arrest people suspected of holding undeclared PoW coins.

    ReplyDelete
  3. The energy usage of proof-of-work coins is clearly a problem and the tendency of the Bitcoin community at-large to dismiss this is also problem that being said I think there is a point to be made about proof-of-work incentivizing renewable energy.

    The problem with renewable energy is we often get a lot of it when we don't need it. for instance we get a lot of solar energy in the middle of the day when the sun is high however energy usage is typically not a tyus them when we need energy is in the evening when the Sun goes down to heat houses run stoves and do other energy-intensive tasks. This makes it hard for a nubile energy to find buyers and is one of the main reasons that we still run so many fossil-fuel plants proof of work can act as a buyer of last resort for renewable energy because most of the operational cost of running a mining farm is electricity cost, miners are very sensitive to changes in electricity price. by acting as a buyer of last resort for renewable energy in x when energy is in excess proof of work can make renewable energy more economically viable and accelerate our transition to it.

    Of course for work only does this good here if we are running on renewable energy sources plus I think a better solution to this problem is a well-made carbon tax this lets us get the best of both worlds: disincentivize in carbon emitting activities and incentivizing renewable energy.

    Typed this with text to speech so I apologize for any 'typos'.

    ReplyDelete
    Replies
    1. Thanks for the thoughtful reply.

      If POW shifted over to renewables, that would be a good step.

      But as I pointed out above in my comment @12:08, even if all blockchain security was 100% based on renewables, it wouldn't change the nature of my argument. Better for gamblers who want volatility to be using non-POW coins that use small amounts of renewables than POW coins that are heavy users of renewables. That would leave more energy for other uses. And the gamblers would be just as happy as before.

      Delete
  4. It's refreshing to see some logic in the digital asset space. These communities surrounding POW remind me of the "The Church of the Children of Atom" in Fallout 3. Which is a group that praises a nuclear bomb in radiated water claiming it to be their savior.

    I think POW should be taxed on all aspects, sales, mining, and annual ownership. This incentivizes all parties within POW to change blockchains, create a new chain, or revise the existing chain immediately. Which is what we need to achieve a more sustainably scalable digital economy.

    ReplyDelete
  5. So you can't convince others performing nonviolent acts to do what you think is best, therefore you must use the threat of violence?

    ReplyDelete
    Replies
    1. It's similar to carbon taxes or cigarette taxes. The idea is to repair a bad collective outcome.

      Delete
    2. Why take the lazy, violent way? Shouldn't you be selling your solution at a lower price, to attract customers? Why use force first, before even trying to come up with reasons persuasive enough to get people voluntarily to come around to your point of view?

      In short, is the collective even worth saving, if it relies on violence?

      Delete
    3. PoS coins are, in fact, already available at a lower "price" (mining overhead) but gamblers aren't switching to them. Hence the nudge of a tax.

      Delete
  6. The whole Proof-of-Work conundrum goes back to the development philosophy of the underlying blockchain.

    In BitcoinCore (BTC) the Core developers have altered the protocols to focus on the digital gold, scarcity narrative. Effectively its a very low-function blockchain. It's coordinated by PoW, and many in the Core community regard (incorrectly) the value of the energy as somehow creating the value in the coin, yet in reality the causality is the reverse. They have a bizarre neo-Marxist Proof-of-Work-Theory-of-Value which is completely false.

    The original Bitcoin protocols were designed to scale transactions and grow the database (as Satoshi noted in early communications). In this sense the PoW is the most innovative, and best solution to the coordination problem. The 'coin' is intended to be a kind of protectionary inflation hedge, but not a speculative asset designed to go the moon! That would clearly make the system dysfunctional.

    This difference in fundamental philosophy is now reflected in the different protocols as you can see that BitcoinSV is doing millions (depending on demand) of transactions a day, growing the database exponentially larger than BitcoinCore, and doing so with the coin at a very low price point, hence low hash-rate, hence low energy expenditure. (e.g. see Coincarboncap.com)

    The energy intensiveness of PoW is therefore not the problem: speculation in an artificially scarce asset 'maintained' on an otherwise functionless PoW blockchain is the problem, and hence the tax and regulatory environment should make the distinction. It's the problem Core created with the "settlement layer" and "layer-2 solution" narrative, which goes against the original purposes of Bitcoin described in the whitepaper.

    Taxing usage based on MWh/transaction or MWh/Mb of data makes more sense. An efficient (non speculative) PoW blockchain will measure up against existing database/cloud providers. But expensive, artificially created scarcity purely as an attempt at regulatory arbitrage by anarchists will need to be taxed/regulated appropriately.

    ReplyDelete
    Replies
    1. Interesting points.

      "...as you can see that BitcoinSV is doing millions (depending on demand) of transactions a day..."

      I have troubles buying that. It doesn't jibe with my (admittedly limited) anecdotal evidence. Are we sure that these are real BSV transactions and not sock puppets making fake transactions with each other to juice their metrics?

      Delete
    2. The transaction count is real. Can see data at say bitinfocharts (which is no friend of BSV).

      Although there are various applications, because BSV is still a little heard of (suppressed) chain, the vast majority of transactions at the moment are coming mainly from one app: CryptoFights by Fyxx Gaming. Can look at data on bsvdata.com

      But the point is, these ae *real* 3rd party transactions with no double-counting (as I understand happens on Solana), and it's like a 10-lane superhighway with capability of 20,000 tps today. Still demand-driven.

      You can check the transactions data in real-time: https://bitcoinblocks.live/

      Delete
    3. "...are coming mainly from one app: CryptoFights by Fyxx Gaming."

      Huh, interesting. Will have to explore more.

      Delete
    4. It's not my thing personally but clearly gaming and e-sports is a huge growth area for scalable blockchains. Fyx (not Fyxx, my error) is an interesting story. Adam Kling the founder started it on Ethereum but it wouldn't scale and was too costly. BSV works perfectly for them: https://www.esports.net/news/adam-kling-interview-on-impacting-crypto-and-gaming/

      Delete
    5. You and I may prefer digital cash to speculative digital gold but why should we impose such a preference on the public? People want to gamble on coins, let them gamble.

      As for BSV, it's not secure and there are PoS coins that use even less energy to deliver the same throughput. PoW is the incandescent bulb of decentralization; it's time to retire it.

      Delete
    6. There are plenty of reasons to regulate money in a democratic society. "Imposing" is a phrase suggestive of a faux-libertarian distaste for democracy, which is part of the problem of the BTC culture. Governments are elected to undertake a regulatory role, not to leave it to one group of society to try and undermine democratic institutions.

      The point about BSV being "not secure" is a non sequitur. This falsely equates "security" with the mob's use of hash-power. Code is not law, and BSV is secure as any legal entity in a functional society where it is illegal to attack infrastructure. If people wish to attack BSV they can do so (briefly) with the full-glare of publicity and expect to be prosecuted for it.

      With regard to Proof-of-stake, the whole innovation of Bitcoin was the use of Proof-of-Work to coordinate a *distributed* (not decentralized) database. Proof-of-Stake is a Rube Goldberg machine for a closely held database organisation. Whether any of these are as efficient as BSV is questionable - Solana being a good example of one that clearly is not. BSV works, cheaply and scales already with no down time.

      Delete
  7. Ok, lets get this shit burn quickly. BTC mining is consuming less energy than Christmas lights in USA…. Fuck Italy, we want to light our houses because….Why? This electricity usage is pure waste. How much electricity consumes banking system? Servers for transactions are nothing in compare with bank houses, transport of banksters etc…. So where is the problem? Other system than PoW is just unsafe and not balanced (premine, big stakes, hacks….) Nobody can hack PoW BTC. Thats why its PoW and will always remain PoW, we can do it eco friendly by using renewable energy.

    ReplyDelete
    Replies
    1. Christmas lights aren't a problem. What we want is people to use the most efficient Christmas light technology, the ones that use the least energy, renewable or otherwise. Likewise, trying to get exposure to extremely volatile coins isn't a problem. What we want is people to be using the most efficient volcoins.

      Delete
  8. An economic philosopher that ignores the usefulness of a commodity and speaks to the (thus) infinite cost. This won't look good on you.

    ReplyDelete
    Replies
    1. Nope, I've addressed the usefulness in my post. When people buy a POW coin like Dogecoin, what they value is Doge's ability to rise by 30% in a short amount of time. I'm suggesting that we push people onto coins that provide this service, but at a fraction of the cost to society.

      Delete
  9. Why is a general carbon tax not sufficient here? Most agree energy production has externalities that need internalizing. Why target one energy use over another? Let’s just tax across the board and let people decide for themselves their updated consumption function. And if it’s speculation you dislike, then why not again a general Tobin tax on short term trades? Or are some short term trades better than others? In general, one should be weary of engaging in a form of misplaced paternalism whenever one aims to institute a selective tax. The burden of proof for instituting a Pigouvian tax should already be high (in terms of evidence for the existence of an externality), and I’d argue it should be even higher for a selective tax, lest one engages in fuzzy moralism

    ReplyDelete
    Replies
    1. "Why is a general carbon tax not sufficient here?"

      The particular market failure that I am suggesting needs correcting is widespread gambling/speculation on expensive POW coins rather than cheaper non-POW coins.

      Now, you can disagree with my premise. But if you momentarily humour me, you'll see that a carbon tax doesn't solve this particular failure. A carbon tax nudges miners onto renewables. But it doesn't shift gamblers off of POW-coins and onto non-POW coins. Even if POW coins are 100% renewable, it's still better for society (assuming my premise is right) to use non-POW coins that are 100% renewable than POW coins that are 100% renewable.

      Delete
    2. JP I entirely agree with your premise, but due to enforcement costs, I think the equilibrium you seek is impossible (this is a 2nd best solution situation). If you'll humor the analogy- it's as if there's a new brand of slot machine that gamblers in every casino love; it's largely the same as other slots, but uses tons of electricity. Yeah, you could tax that brand of slot machine to raise the price. But if they generate tons of gambling revenue, then casinos will still buy them.

      The analogy is imperfect, but reality of coin-gambling only seems to suggest an even more expensive enforcement case. I don't think there's likely to be a cost-justified solution to push the gambling demand elsewhere to correct this market failure. But I would love to be convinced otherwise.

      Although a general carbon tax is not sufficient to solve this particular market failure, it is (theoretically) sufficient to solve the aggregate costs imposed on society by similar market failures (broadly: energy efficiency externalities, recognizing this is an extreme case).

      Delete
  10. Probably the easiest and most straightforward option I can think of would be a crackdown on larger scale mining operations. Either tax the electricity they consume on the utility end, or just enact more punitive measures on mining directly. These specific operations are not particularly resistant to state power as its quite easy to find them by doing some detective work on power consumption. I don't really think that enforcing taxes/etc on smaller miners is worthwhile given the comparatively paltry sums and greater difficulty of discovery. Trying to implement taxes at the point of conversion to fiat also strikes me as quite impractical, as I am dubious that the federal authorities would realistically have a handle on the wild west of pure crypto to crypto exchanges. Tracing hundreds of different blockchains interfacing with each other seems like an incredibly inefficient use of enforcement resources to prevent people from laundering PoW coin gains into PoS coin profits.

    While this would still leave some smaller scale miners with a GPU or so untouched, it would certainly be able to lower the hash rate, potentially low enough that a 51% attack could be conducted which would cause a price collapse and a loss of interest in speculative activity.

    ReplyDelete
    Replies
    1. Thanks Roth, great points. I like your reasoning: a higher tax on large scale mining leads to reduced hash rate that reduces security. This either nudges speculators off of POW coins or convince them to pressure POW coin developers to make a switch to PoS, since a switch would increase the price of their POW coins.

      Delete
    2. Besides taxing at the point of energy consumption, you could also implement taxation or crackdowns on the Bitcoin ASICs used to mine crypto currency. While there are other PoW coins that use GPUs or CPUs, most PoW economic activity is done with Bitcoin. In addition, there's other non-Bitcoin cryptocurrencies that use the same mining algorithm or merge mine, so you'd have pretty significant impact on the crypto economy as a whole while having pretty much no impact on other uses of computing resources, and potentially even positive supply side benefits if Bitcoin ASIC manufacturers decide to shift production to more utilitarian semiconductors instead.

      However, I do feel that it is worth noting that there are risks associated with such policies. For starters, Proof of Stake currencies don't really have security that's as strong as PoW because of various technical issues such as the Nothing at Stake problem. As a result PoS security is typically highly reliant on quasi-institutionalized consensus from major shareholders of coins or even de facto total centralization, thus it doesn't have the same degree of fairness baked into the system. The primary reason Ethereum hasn't switched over to PoS yet is because the developers recognize this problem, and shifting over now could have potentially disastrous impacts on the stability of the blockchain.

      Therefore the demand for PoW coins will remain because they have inherent properties that PoS coins don't, so the possibility remains that by cracking down on Bitcoin/PoW in general, you end up with a large amount of liquidity pumping into more illicitly inclined PoW cryptocurrencies like Monero rather than it all going into PoS as desired. While such an end result would be far closer to Satoshi Nakamoto's original ideological vision in some respects and would be quite pleasing to many crypto enthusiasts, I doubt entities like the IRS or banking regulators would find this outcome preferable. On the bright side, it would probably still be significantly more environmentally friendly since RandomX is more efficient so I suppose it depends on how you prioritize matters.

      Delete
    3. "...such as the Nothing at Stake problem."

      Isn't that addressed by slashing?

      I confess that I'm not informed enough to weigh in on the PoS/PoW debate. As an outsider I sort of assume that because so many chains use PoS (and major chains like Ethereum and Doge want to switch), then PoS has attracted sufficient consensus from blockchain technological elites that it is sufficiently secure.

      "Therefore the demand for PoW coins will remain because they have inherent properties that PoS coins don't, so the possibility remains that by cracking down on Bitcoin/PoW in general, you end up with a large amount of liquidity pumping into more illicitly inclined PoW cryptocurrencies like Monero rather than it all going into PoS as desired."

      I agree that there's some artisanal demand for POW that's always going to be there. But I don't see why a mining tax would drive Bitcoiners into Monero. A tax penalizes all POW coins equally, no?

      Delete
  11. "....then PoS has attracted sufficient consensus from blockchain technological elites that it is sufficiently secure."

    Prices are not determined by "bitcoin technological elites" but rather a combination of speculative trading from uneducated investors as well as insider trading from more organized entities such as mining cartels, cryptocurrency exchanges, and various public opinion manipulators. Chain consensus for pure PoS coins is maintained by either single centralized entities or a small cartels who are able to maintain 51% control over a PoS blockchain at zero cost. As you have mentioned before, the vast majority of cryptocurrency usage is basically gambling, and setting up a framework for such activities is certainly something that doesn't inherently require the absolute decentralization that PoW provides.

    "A tax penalizes all POW coins equally, no? "

    The primary reason why large commercial mining entities emerged at all to begin with was the emergence of specialized Bitcoin ASIC mining hardware which are over 1000 times more efficient than consumer grade hardware. Individual miners would not be able to compete with the economies of scale that larger scale ASIC mining operations would provide, and hackers utilizing cryptojacking would get paltry returns on effort from hijacking commercial hardware not dedicated to Bitcoin mining. However what makes Monero different* is that it uses ASIC resistant RandomX as its mining algorithm, which means consumer grade CPUs are the most efficient means of mining it. As a result individual miners and hackers naturally out-compete large scale commercial mining operations because they have significantly lower capex and energy costs.

    A single hobbyist mining cryptocurrency can repurpose a CPU used in day to day gaming instead of having to buy dedicated hardware, and its much easier for someone at that level to repurpose the miner to act as a space heater during winter or gain access to cheaper forms of subsidized power. Hackers using cryptojacking software to mine cryptocurrency obviously don't have to pay for hardware or power. Therefore a mining tax would be much more likely to penalize cryptocurrencies like Bitcoin because they are reliant on large scale mining operations for hash rate while Monero and a few others are not. You could *try* to crack down on individual mining activity but it'd be significantly more difficult to stamp out.

    At the end of the day, its worth noting that when considering the inherent limitations of the US political system (1st amendment protection precedents, technical incompetence and gridlock in congress, etc) the most effective institutions in regulating cryptocurrency activity are the miner cartels and centralized exchanges themselves. If it wasn't for their swift actions keeping Monero from deeper pools of fiat liquidity (made in self interest of maintaining their insider trading revenues), cryptocurrency as a whole would be a far greater threat to global economic stability due to the potentially greater capacity to conduct tax evasion, capital control bypassing, and money laundering with it.

    *(Its still theoretically possible to use ASICs to mine Ethereum profitably, but the somewhat ASIC resistant algorithm in conjunction with future plans to transition to PoS have largely discouraged ASIC manufacturers from investing in manufacturing capacity for it. In addition ASIC manufacturers typically don't bother making ASICs for PoW cryptocurrencies with market capitalization that is too low. There is also Proof of Storage, a close cousin of PoW that is technically ASIC resistant but I am unsure if it has the same degree of resistance to dedicated mining hardware in the long run.)

    ReplyDelete
    Replies
    1. Apologies for the addendum, but I have a few notes I'd like to add that I forgot to include.

      "and major chains like Ethereum and Doge want to switch"

      As I have mentioned "true" PoS is de facto heavily centralized in order to maintain security or piggybacks off of PoW security (i.e. variants using Proof of Burn). While the power savings are appealing, what keeps coins like Dogecoin and Ethereum from swapping over immediately is that they're still trying to figure out how to do PoS without the establishment of semi-centralized institutions other examples utilizing PoS have had in order to maintain chain consensus and security.

      Delete
  12. The amount of time spent worrying about what amounts to a rounding error of global emissions (0.3% in a worst case scenario of 100% coal-powered mining, according to CBECI) is fairly staggering. A tax would literally accomplish nothing. The environmental effects are vastly overstated and simply do not have a meaningful impact on the environment.

    ReplyDelete
  13. But without POW we have cryptos that are all securities right? How else do you create them out of thin air in a decentralized way so that when you go to sell them they weren’t issued from you instead you mined them,
    Also the government gave a green light to pow - and miners - they aren’t selling securities and they aren’t money transmitters.
    Even if you taxed pow coins which I don’t agree with pow let’s new pow coins throw off the sec since they are mined.
    Also you can become an Enron energy trading whale and manipulate energy grids with pow coins

    ReplyDelete
  14. I read somewhere that bitcoin mining uses 127 terawatt hours a year. Forget Italy - that's more than Norway. At this point, it's not going to be worth it - at least at home - unless you're using solar power or some other renewable energy. I mean, at what point are you spending more than you're making? I have to imagine that the payoff will become less and less as more and more people keep mining for it. And especially when someone's programming whole server farms and whatnot to do it. Suddenly it's not as viable for an individual. Plus I spoke to an electrician friend of mine who said he's been upgrading panels and lines into homes so they can handle the load with tripping breakers. He's on the fence about whether it's worth the investment for the person doing the mining (although of course it works out just fine for my friend, haha).

    ReplyDelete