Hacker News new | comments | ask | show | jobs | submit login
Valuing Cryptoassets from the Ground Up (medium.com)
38 points by msall 9 months ago | hide | past | web | favorite | 18 comments

But are cryptoassets distinct enough to be considered separately?

So to take your example of gold, if BTC is gold, then is ETH more like silver or more like a new source of gold? If ETH is silver then there will be some base demand for ETH separate from BTC but if ETH is just more gold then all we've done is drastically increase supply. To me it seems that demand for specific cryptocurrencies isn't really that high while demand for cryptocurrencies in general is, that makes ETH and BTC seem both like gold to me and that makes me question the fundamental notion that the appreciating cryptos have supply restrictions.

I've always struggled with this idea that people are trying to apply logical models to illogical situations. Ideal buying size is not going to save you from a 10 minute pump and dump.

Optimal holding percent is not going to protect you against a total whale coming in and selling off because they're bored of the currency.

Trading is based on emotion and the control of that emotion. Buy into technology you understand and find valuable. The only calculation you need is to know what you're willing to lose and set your stop loss, everything else is a bit of a waste in my opinion.

> ETH executes all VISA payments

> ETH Value: $0.088

(There are 100M ETH)

Correct me if I made a mistake, but it seems like this is not a stable market. What's to stop me from buying a large chunk of the supply and sitting on it? That would increase the value of my ETH because VISA throughput would have to be maintained. Many others would come to this same conclusion, driving the price up massively.

Right, this would place the value of the ETH network at $8.8 million if it were processing all VISA payments. Something is wrong with the author's math or conclusions on that point. The value of processing as many transactions as VISA should be self evidently worth billions.

Another factor in crypto is that especially with proof-of-stake coins (like ETH is heading toward), part of the security of the protocol is in the expense of doing a hostile takeover of the network. A paltry $8.8 million valuation would presume that someone with as little as that could buy up all the supply (or 51%) of ETH and take effective control of the network.

You're right that attempting to do so would drive up the price, very rapidly.

Another sweeping point the author makes without backing up is the idea of "infinite competition" in the crypto space. Yes, anyone can fork BTC or ETH and make their own coin, but miners, stakers, exchanges, users, app developers, etc, won't necessarily flock to it. "Trust" is not built into the protocol.

I have no faith that someone who clicks the "Fork" button on Github necessarily has any idea how to run, maintain, secure, and grow a blockchain into doing something unique or novel. And I wouldn't stake my business or assets on it, per se, unless I was convinced that a team of people and a community was able to spring up around that blockchain.

Network effects matter, especially in crypto-assets. I trust the BTC network, because it's gone 9 years without a major exploit. I trust it because of the community, network, and ecosystem behind it. To act as if a fly-by-night crypto can swoop in and steal that crown so easily is foolish. Anyone can make a Linux fork, doesn't mean you're going to dominate the desktop market.

It's a good point that the low market cap could lead people to try to buy a controlling supply. I explain in the other comment why I think it wouldn't be profitable on a pump-and-dump basis. On the basis of control and trust, maybe there still needs to be some assumption that at least 51% is being held by various parties, enough so that the market trusts the network to not manipulate. I need to think about that more. I'm really glad you brought it up.

As for network effects, I find that to be a super interesting question. Because you're right, there are definitely certain levels of trust that existing networks have earned, and there's a good argument for why that creates friction and earns those networks the ability to charge a premium. I personally don't think that extra friction is long lasting. I'd expect that if there were indeed a new solution that accomplished the same thing at half the price, it would get attention and the market would gravitate to it fairly quickly (maybe not within weeks or months, but years? maybe it depends on the timeframe you want to consider, and the level of improvement it provides). But it's a really good point, and something I think about a lot.

If you look at it from a utility perspective and discount the network effect, at a certain point, it becomes economically incentivized sharding.

"We accept any of the 2021 or prior World Bank approved ETH forks." Your multi-wallet client automatically chooses the one with the lowest transaction fees (also economically regulated).

The author is assuming that (a) someone who's making all their ordinary purchases with ETH instead of VISA isn't holding any ETH, but making a separate buy on an exchange for each individual purchase, and (b) anyone who receives ETH immediately sells it for some other currency, despite living in a world where it's possible to use ETH for all purchases. These seem like questionable assumptions at best.

There's an underlying assumption that a cryptocurrency must be either a store of value or a medium of exchange, but not both, which is a bit odd since every fiat currency is both, and gold was both for thousands of years.

That brings up a good point I should make explicit, which is that I think those two sources of value are separate but not necessarily mutually exclusive. I think you can consider them individually and then add them up. John Pfeffer makes this argument too, that the two sources of value likely don't affect each other too much, allowing you to consider them separately.

That makes some sense, but I think there's also significant friction in the conversion. If I'm a store owner getting paid in ETH (or whatever), and I can buy a lot of my supplies with ETH, then it makes sense to just keep the ETH and buy my supplies with it, instead of paying unnecessary exchange fees. And if people do that, you don't have super-high velocity anymore. You're just an everyday currency like dollars (which are also mostly digital these days).

In fact, if people were to contact an exchange for every transaction, it's hard to see how there'd be any advantage over just using legacy financial systems.

I used to put a lot of stock in velocity of money considerations but now I'm thinking it doesn't have that much predictive value. All it's really saying is that the GDP is defined as the number of transactions, times the nominal amount of those transactions, times the real-world value of the currency unit. The currency price could be anything and the equation still holds true. If you assume a maximum achievable velocity, you can work out the minimum currency price for a given GDP, but the price could also be arbitrarily higher than that.

That's a good point. The main reason for the low price is that velocity is so high — all of the coins are changing hands every 10 seconds in this scenario. So there is still a huge amount of demand for them. If you tried to buy a lot to drive up the price, I think all of that demand would be recalibrating the price so quickly that practically speaking it would be hard to sell them back fast enough at a price any higher than you paid. So I think it would be hard to manipulate the price in a profitable way.

But it's a good point — maybe there's some minimum level of holding that happens (could be high, 99%? 99.9%) that is solely about making it expensive for others to increase the price arbitrarily (even if they can't do so profitably), in order to maintain price stability. But that would be about paying for stability, not for the underlying utility. I need to think about that more.

I see what you mean about not being able to sell back into a different currency profitably. I need to think about this a little more myself.

On a slightly different note, another relatively minor but important thing to take into account when valuing from fundamentals is the percentage of permanently burned supply. ETH may have a supply of 100M in theory but in practice, maybe it's 70-90M due to private key loss or smart-contract lock.

What about possible protocol-enforced scarcity/supply-side limits even in the potential future "mass adoption" scenario?

I see cryptocurrency as a harbinger of AI, which is the real revolution. Blockchain is just a tool. People get really confused trying to understand and develop pricing models for the current state of the world but I see a lot of cryptocurrency value as reflecting the adoption of AI.

The author's equation produces low valuations at high velocity and this suggests a use case for low valuation, high velocity cryptocurrency as a medium of exchange if not a deflationary store of value. I don't think this currency exists yet, though I get lots of pitches for things that want to be this while also trying to justify their own value into the future.

In the early days of crypto, libertarian economists predicted that bitcoin would go to the moon and the USD would undergo hyperinflation. This is clearly not happening. What I think could happen is a relatively stable USD under the guidance of the Federal Reserve and a simultaneous hyperdeflation of ETH and BTC as they become the reserve currency of distributed AI capital organizations.

> I see cryptocurrency as a harbinger of AI, which is the real revolution. Blockchain is just a tool. People get really confused trying to understand and develop pricing models for the current state of the world but I see a lot of cryptocurrency value as reflecting the adoption of AI.

This seems like a strange thing to believe. Why do you believe it?

I don't see nearly as much use case for humans and cryptocurrency as for AI and cryptocurrency.

I think in the future, the average technologically savvy person will be just as confused about what bitcoin is and does as they are today and it will mostly be used to securitize options and derivatives of unprecedented complexity.

Bitcoin is pretty bad at offering advanced securities or derivatives.

Really bitcoin is pretty bad at everything except atomic transfers.

Bitcoin is shit at loans, it's shit at writing wills, it's shit at scheduling payments, it's shit at offering shorts, it's shit at offering options, it's shit at providing oracles etc.

Right. I imagine a future where AI capital entities just hold bitcoin and never do anything with it other than hold it forever to secure the things they're doing on the advanced financial platforms of the future.

If the only action your AI takes is to hold bitcoin I've been running an AI for many years.

Capital that you can not leverage is a disadvantage.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact