Skepticism re The Crypto Utopia

Six problems I have with the mania today

Occasionally I see pronouncements about how we'll all live an on-chain world very soon. There are plenty of bitcoin billionaires now, more than a few bitcoin millionaires and an incredible array of evangelists. There are fantastic threads filled with the wisdom of the converted, deep in the excitement of rediscovering economics insights from the past centuries. But I’m skeptical.

For a taste as to why, the chutzpah of this tweet is pretty dazzling. His answer was 10 years, give or take, by the way. For the dollar to stop living. But regardless of what your exact definition is, if someone wants to bet here, you should probably take it!

I'll start by saying I'm willing to change my mind about any and all of the following, and I have held a bit of ETH I've held since 2017 or thereabouts, which has cured (some) of my FOMO. So if you do invest in bitcoin, more power to you.

And a lot of the arguments however seem to flip between intriguing and exciting and economically illiterate, which piques my Gell-Mann Amnesia. But even so, this phenomenon has had exceptional hype from those who have gotten rich off of it, with zero indication that they think talking their own book isn't particularly unbiased, and that making money is not the same as leading a social movement.

So here's a look at some of my enduring objections re Bitcoin and it's utility, the Exchange medium argument, the Monetary store of value question, Inherent value argument, a look at DAOs/ DeFi and NFTs. I hope someone can answer them!

Non-fungible Bitcoin as a unique global currency

One of the key benefits of the digital age is cost-free reproducibility. That's what Kevin Kelly wrote about as better than free. His thesis is that with the advent of high compute power and the internet, information is essentially free. Therefore:

When copies are super abundant, they become worthless. When copies are super abundant, stuff which can't be copied becomes scarce and valuable. When copies are free, you need to sell things which can not be copied.

The key issue with a digital only currency that confers extraordinary wealth to those who bought it early due to forced deflation, is that everyone who comes late to the party has an incentive to do a Bender, and create their own currency, make it easier to use and adopt.

Why would you continue to make Satoshi richer rather than make your own Satoshi? Wouldn't enough people do it such that a new equilibrium is formed? It wouldn't even need to be perfect replicas, though of course it could be. In fact we've arguably started seeing it too with Bitcoin Cash, Ethereum original flavour and others.

While of course there's friction, if we're attacking a multi trillion asset class, I'm not sure why this won't happen, and happen with the endorsement of big celebrities who for whatever reason didn't manage to cash in on the Bitcoin craze the first time around. If Kanye starts selling his next album for KitCoins, a copy of Bitcoin but with a K, I'm guessing some demand will be generated.

The answer so far seems to be that we've all accepted Bitcoin as the reserve dollar equivalent, implicitly or explicitly, due to its long tenure and precedents. At which point I can only shake my head and point to the Dollar. Or the euro. Or the pound. Or the ...

If all currencies are digital, as of course most of them are, then by definition it's a competitor to crypto. If the problem is that currencies are "backed by the faith and credit of the government" while crypto is "backed by the faith of those who hold it", I'm not sure that's a cagematch crypto should enter, frankly.

Mostly therefore, to put it bluntly, isn't this just a Greater Fool Theory in action?

Don't exchange mediums generally require much more scale?

By its very nature, once Bitcoin is super expensive, using it as a medium of exchange becomes extremely expensive too. (Unless you believe it'll all be in fiat and Bitcoin wil just be an exchange mechanism in which case, welcome to my objection? You've essentially discovered a cheaper Western Union). If Layer 2 solutions don't work, then it can't work. And while transaction volumes are ever higher, they're still miniscule in the global scale, though of course this is true at some point in time for all new transaction mediums.

According to crypto market data aggregator Messari, the 30-day rolling daily average for Ethereum transaction volume is currently $7 billion, with Bitcoin processing less than $3 billion.

And if you assess the transaction volume estimations in USD vs the confirmed transactions per day we can see a couple of trends. One is that the transaction volume seems to spike towards the end of 2020, while the confirmed transactions per day stays pretty steady, showing that almost all the changes came about because of price increases.

I'm not sure if this is indicative of a high degree of ongoing usage. Of course things can change, but it's worth noting Visa can theoretically handle 65k transaction messages a second, and $11 trillion total volume in 2018. Now Ethereum can handle 15 transactions per second (TPS) compared to VISA's 1,500 TPS.

All I'll say is, there's some work to be done here.

If the argument of crypto about immutability and cryptographically recording every I/O transaction through computationally intensive (and therefore tough to break) calculations is true, then by definition it's an extraordinarily environmentally expensive endeavour.

You could scale back the proof needed, as indeed some are doing, but at that point aren't you just building the next Visa or next Stripe? It's cool, to be sure, but this isn't quite the same as Muggles discovering magic!

It's the new monetary store of value

Digital gold!

For this to work everyone has to admit that everyone else loves BTC and therefore they should too. This is the Keynesian beauty contest turned up to eleven. Will they? Maybe it'll become the reserve currency of the crypto economy and probably part of our overall economy, because gold is boring anyway.

When the evangelists say BTC is the best because everyone has access to it, that's just inherently ridiculous. A much fewer fraction of the world has access to it than their local currency or investing methods. While it expands market participation from some external countries though only through non approved brokers or exchanges which have the exact same worries as fiat, it contracts it from those who are not as tech savvy which is also a massive majority of the world.

There's also an argument here that Bitcoin is rising because of implicit inflation worries and it's a better store of value. One version of the argument also talks about M2, it's changes, and how Bitcoin earned its trust with the spike.

Of course this doesn't make much sense. I did three things to check (across 2 time horizons - 5yr and 2yr) - 1) added M2 alongside M1 to include the savings, 2) added Nasdaq and TSLA as other assets, and 3) did a correlation matrix amongst all of them - https://docs.google.com/spreadsheets/d/1tQK2rEOnbi5F-1uuSNkPcUP5AUUls25OPNeaUqt9RmY/edit?usp=sharing

To make the case for asset reallocation based on M1/M2 btw you'll also have to look at the changes in absolute values. For instance (back of envelope math ahead), since 2016 BTC has 10x-ed, so added c.$900B, while M1 roughly 2.5xed, adding c.$3.7 Trillion. Same time Nasdaq added around 5 trillion in market cap (though not cash inflows).

From a quick look it's not worry over M1 that's driving BTC, but rather the fact that there's clear cash inflow into sources of yield, including BTC, Nasdaq, Tsla and others. If you wanted to expand I'd also add interest rates (https://fred.stlouisfed.org/series/FEDFUNDS) though that one peaked around 2019 then fell off a cliff so is muddier.

Also, if deflation is per se interesting, then making money supply restricted has clear impacts. For one thing we're giving up any hope of inflating the money supply at times when we really really need it. Like with recessions or depressions or world wars or pandemic responses.

In reality, deflations suck quite a bit. They aggravate recessions and reduce economic growth. They even cause a spiral, one that's not much fun. If deflation comes because bitcoin effectiveness/ productivity goes up, then it's okay, but that's not even close to guaranteed.

For one thing, it also prompts hoarding rather than using (see Chamath's $127mn spot of land). However if we only use Bitcoin and it's ilk as an exchange medium, then of course there's no problem with inflation or deflation, since it's only a conduit, and as long as there's no intra-transfer volatility in price, the actual price value doesn't matter.

The inherent value argument for Bitcoin

The cheekiest way to do this is to assess BTC impact as equivalent at least to the cost of mining. But this is likely to piss everyone off and start even crazier comparisons to BTC as a battery. As amusing as that is ...

(the crazier option if you're Elon Musk is to start your own mining operation so your batteries fight your Bitcoin which ends in an orgasmic level of cosmic irony)

While there's an argument on "why not mine Bitcoin rather than waste peak time electricity", this isn't instructive to the value argument, since that's just the marginal price of electricity. Or maybe you can do an anti-BTC portfolio here and develop batteries which, you know, actually store energy!

There is a wonderful analysis done that tries to analyse Bitcoin value according to quantity theory of money. It states:

Bitcoin Value = 1/P = T/(M*V). Where:

  • M is the money supply

  • V is the velocity of money in a given time period

  • P is the price level

  • T is the transaction volume in a given time period

The velocity of the United States M1 (highly liquid) money supply hit a high of over 10 in 2007 and is now around 4. The velocity of the United States M2 (moderately liquid) money supply hit a high of 2.2 in 1997 and is currently at less than 1.5. Currently, the velocity of Bitcoin is much higher on average, but the problem is that a large portion of this velocity is just trading volume, not spending volume.

Which gives us this beautiful comparison table.

Another option is to think of BTC as comparable to a national currency.

If there are 8 billion people in the world in ten years, and 5% of them use Bitcoin, that’ll be 400 million Bitcoin users. If the average Bitcoin user does only 10% of their economic activity in Bitcoin and 90% of their economic activity in typical currencies, then that’s the equivalent of 40 million people using Bitcoin for 100% of their economic activity, or roughly the size of the Canadian economy assuming similar average per-capita economic activity.

If Bitcoin’s reasonable market cap becomes worth, say, $1.5 trillion in that scenario (comparable to Canada’s M2 money supply), and there are 20 million bitcoins in existence by then, each bitcoin would be worth $75,000.

Compare to Gold!

All the gold in the world is worth maybe $10 trillion, based on the World Gold Council’s estimate of how much gold has been mined and what the per-ounce price is. In other words, maybe 2-3% of global net worth consists of gold.

If Bitcoin’s total market capitalization achieves half of the global value of gold ($5 trillion, or about 1-2% of global net worth) and the number of bitcoins at that time is 20 million, then each bitcoin would be valued at $250,000

There's also a stock-to-flow model that analyses BTC price comparing it to the amount that actually gets circulated.

There are several other methods, each more arcane still, which might be correct, though the jury is very firmly out on them!

The key point however is that all of these are academic at the moment, with limited actual "quasi-GDP" style data to help inform what to choose. We could debate odd theories and we should, but it's hardly conclusive.

DAOs/ DeFi

Programmable chains will have the impact that we will be forced to specify legal systems to a code-like level , which will naturally bring forth Hopefully it increases the level of precision within companies and the level of experimentation within company governance structures (this is a great twitter thread on it).

That said, we can barely experiment with the structures we have.

While this is true, it's worth pointing out that calling an open source project a startup and ascribing success is a little disingenuous. After all, even before this we had Linux, the OG of the open source world, and hundreds of projects besides.

My big question here is that in most instances, including Ethereum, we have to deal with people. And people are notoriously unreliable in their willingness to be put into strictures or boxes, much like the world itself.

After all, we've had lawyers for centuries trying to nail down normative human behaviour into codified language, but I'm not sure that is a project we can call a success. Much like that DAOs will help, like any smarter legal contract will help, but the messiness is not because of incoherence of information, it's because the world is inherently messy.

Even if it's through the creation of a new coordination mechanism so we can run companies together like voting enabled open source projects, getting it done still faces the same obstacles.

And re DeFis, similarly, not having gatekeepers is great for increased efficiency and reduced interference. But not having gatekeepers also sucks for resolving any corner cases, which is the biggest issue with any highly specified systems. The reason I need AmEx is not just to help me buy my coffee, it's because when fraud inevitably occurs I get to call someone and reverse it.

Ultimately the problem here lies in the fact that these systems do make transparency easier and friction costs lower for audits, but mainly if you play entirely inside their gardens. And that's true of all tech. If you're able to codify your organisation and run all decisions and processes and cash management through one application that would be awesome indeed. If.

NFTs

And finally, NFTs. See the difference between a da Vinci painting and it's replica is that the replica isn't a proper replica. The brush strokes are different, the paint used is different, it's a different painting!

What if there's a perfect forgery? Is it worth the same? If you ask the buyers, they say if you know it's provenance the whole qualia is different. Which, since I don't particularly hold to the hard problem of consciousness, feels like a cop out.

On a related note, turns out people love autographs! And you can even invest in the autograph market.

“None of my other investments give me the joy that autographs do, because they make me feel that I am holding a piece of history in my hands. In my view, they are the most undervalued of all areas of collecting, especially when you consider what people are paying for even second-rate paintings.” ... So confident are they in the rising future value within the autograph and stamp markets, they are offering investors an interesting deal. A deal where your upside is unlimited and your downside a minimum guaranteed 15% return after three years or, for those willing to commit to a longer time horizon, 70% return after 10 years. So you can’t lose.

Also did you know that there's a thing called the Autograph Index?

The PCF40 Autograph Index, which tracks the performance of the 40 most regularly traded autographs in the world, recorded a 10% annual compound increase in value between 2000 and 2020.

Why are autographs worth anything? Because people pay for them. It gives the feeling of being closer to fame. Which is why these are the most expensive autographs sold in 2021:

Still nowhere close to NFT prices.

So if Naval replies to my tweet, and I sell that, will that give the buyer a similar feeling of being closer to fame? Maybe. A baseball card is fungible until it's signed. But even when signed it's not assigned to you. If I buy that baseball card I can believe that it's the card + the signature that causes the value. But if you buy Naval's reply to me, our identities are inextricable.

That's why NFTs make no sense to me. They're:

  • Forcibly making a fungible item non fungible by adding an encrypted marker

  • The artwork made first vs it's 100 copies are indistinguishable. They're perfect replicas.

  • Autographs aren't independent entities, they're made on already interesting things with a value. There's a reason you're not selling Babe Ruth's scribbles in a random notebook, vs jersey. Though even his baseball was pretty cheap by today's standards!

It's a combination of some intrinsic value + some future value because the artist will become famous + some future value because the artwork itself will become famous.

So while it makes some sense to buy a piece of digital art because "SHL0MS made this themselves and I can prove it", it's also worth entirely that because of perception.

And you know what? I get it. People but the craziest stuff when they think it'll be cool one day. Wouldn't you buy JK Rowling's first email conceptualising a dragon? I mean, no for me, but there's a market!

What I find most fascinating about this is the nihilistic viewpoint. If money doesn't matter, and fiat is the same as worthless paper, then why not throw $60 or $60 million to buy something random. It's all just ... bits and bytes in the eye of the beholder!

A glance at the arguments in favour

So I don’t get stoned, here’s an overview of why I’m (cautiously) optimistic.

  1. There's a regulatory arbitrage opportunity because if you want to open an account with just a phone from anywhere in the world today, you can't do it. Self custody wallets can probably solve this problem. I love this idea as an on-ramp to financial system. I also love the idea that this is a way to bypass/ control/ restrict/ enhance the existing KYC or AML systems, which generally suck!

  2. I love the idea that you can embed cryptographic information into messages and ordinary transactions. It seems like an evolution of the existing transactional systems where concepts like proof and verification are part of the message. On a philosophical level, in any message there's an implicit piece of information about who the sender is. Provenance matters. This is a way to embed provenance into communication.

  3. Governance kind of sucks in most organisations. The decisions are made by management with some oversight from the Board and some through the voicing of shareholders. Bringing in the community has worked quite well in the open source community, and the ability to combine the community governance aspects of open source with the running of a company has some value. While I don't think it's something that can work at scale, since coordination taxes are real, it adds a new method of governance that could be really quite cool. The joy of the digital world is that we can coalesce opinions at scale, so might as well try it!

  4. Distributed and federated systems are fault tolerant and resilient, maybe even higher performance. I started my coding career learning from BitTorrent and P2P networks, and since then I've been a fan of using complex coordination logic to solve problems that you can't through pure selection. For instance, downloading a file 1 to 1 takes time that's dependent entirely on the lower of yours and their bandwidth. But once multiple people want the same piece of information, suddenly you can create a network where individual file pieces are not restricted to the lowest possible bandwidth of one sender. That's important. A lot of the problems in the world comes from our inability to handle coordination issues, so this could be a massive step forward.

  5. Back-ends for most software systems suck. Partially because of legacy tech debt, but also because they're often held together with duct tape and hope! There's a reason Oracle still exists as a behemoth, because nobody can rip out an ERP that easily. Crypto gives at least an option for existing solutions, whether piecemeal like an individual database or all in one, to start knitting together and create immutable systems of record. That's pretty cool in itself!

  6. It could decrease transaction costs, because as a new form of exchange, getting everyone in the world a crypto wallet might be easier than getting everyone a bank account. Also, microtransactions could be awesome. Is it a psychological cost, or is it an economic cost, to pay for things in increments of $0.0001? It’s unclear. While I'd love to see this happen, it's unclear why we need blockchain for this. Theoretically everything is divisible, so you could just as well say I take payments in increments of 0.01 cents, but only pay out once it hits a threshold, just like Starbucks coupons.

The future of crypto might be quite bright. But I'll wager that in 10 years what we see or use as crypto will have little similarity to today! It's a bit like folks trying to cash in on creating TCP/ IP, rather than building stuff on top like the internet. Or people trying to cash in on the internet without actually building applications. Information superhighway is a nice catchphrase, but those highways need rest stops and retail shops to make stuff.

Mostly though, now the money's almost entirely made by the early holders. Which isn't particularly conducive to it becoming a "movement" much less a populist one. If you want people to use it, you might want to stop shoving the fact that they're dumb for not having bought it in 2010 down their throats. As a rule people usually don't like being made to feel like fools.