Ask HN: How do we know Bitcoin is not in a bubble?

28 points by tbirrell ↗ HN
Bitcoin's rise has been astronomical in the last year or so. How do we know that this isn't a bubble that will bankrupt 95% of its users?

54 comments

[ 3.9 ms ] story [ 66.4 ms ] thread
we don't.
We even know it is a bubble.
As a fiat currency, it is almost by definition a bubble. It's fundamental value is essentially nil, yet it's "price" is positive. In the current equilibrium, people seem to agree that Bitcoin is a good store of value (because it is perhaps limited in supply and ownership is verifiable). Fortunately/unfortunately, other cryptocurrencies are just as useful in this regard. As such, Bitcoin's value is something of a sunspot equilibrium. Any sufficiently big shock or event could shift the focus to another cryptocurrency or other product entirely.
"It's fundamental value is essentially nil,"

That's not strictly true. It's fundamental value is the effort required to obtain it - which at the moment is some rather large amount of electricity and hardware. If you have deployed those resources you will not let the Bitcoins go for any less in real exchange unless you absolutely have to.

That is the same for anything that is a currency. It's base value is what you have to do to get it.

The problem with crypto-currencies is that nobody has to hold them to deliver them anywhere. There is no solid drain other than to savings.

You've confused fundamental value with fundamental cost. A hand-knit sweater will not necessarily (or even usually) fetch a selling price equal to the input value of the labor that went into it.

There are many endeavors which produce goods whose market-clearing price is below the value of effort required to produce them.

Houses are a good example. When land is plentiful, the cost of a house is usually a slight discount over what it would cost to build a similar one for new.
Your first paragraph basically describes the sunk cost fallacy. People have put money in, so they will make sure the value doesn't drop below their investment.
This is the labor theory of value and proven wrong. All value is subjective.
Bitcoin is technically not a fiat currency. Fiat currencies are established by governments and are inflationary in nature since the governments can print more money whenever they feel like it. Bitcoin is deflationary and more like digital gold since the supply is limited.

Bitcoin could end up a popped bubble if Tether[1] loses adoption.

[1] http://fortune.com/2017/12/05/bitcoin-btc-price-usd-tether-l...

It probably is a bubble.

The problem is when it bursts, we don't know if we'll be in better positions having entered now, or having entered after.

It very obviously is. Nothing can grow at such an insane rate and be a stable, long-lasting thing. No one knows how high it will go, but that it will collapse hard soon is 100% sure.
If you are certain it will collapse you can make a lot of money shorting it. If you are 100% sure, there is no downside.
Generally the problem with shorting is that the shorts are time limited - because it is a derivative bet, not a proper loan.

And following Keynes the market can stay irrational longer than you can stay solvent.

What financial mechanisms exist to short Bitcoin?
Here's an idea. Ethereum smart contracts to short bitcoin. You borrow my BTC for 30 days, with a 5% fee. Someone could make a killing off of that idea.
I used to short/long bitcoin/eth on GDAX, you have to register as a business and turn on Margin Trading
(comment deleted)
Maybe you've heard the old finance saying: "Markets can stay irrational longer than you can stay solvent".
Is there even a US based legit platform to short Bitcoin? If Coinbase was serious about having a true market they would allow you to short Bitcoin as well as buy.
CME is offering bitcoin futures starting next week.

Interestingly enough, S. Korea has declared bitcoin futures illegal today and S. Korea is the leading buyer of BTC.

Should be a good time.

Coinbase does have a market tho, it's called GDAX, you use your coinbase login and it has margin trading on it. Once I found out about GDAX I never used coinbase again for trading currencies because GDAX has no transaction fees if you are creating the transaction ie. setting up the buy/sell limits
> Nothing can grow at such an insane rate and be a stable, long-lasting thing.

That sounds like common sense, but common sense is not neccessarily right.

Imagine a German in the 1920s looking at the price of US dollars in Reichsmark. That can't be true, $1 for 1000 RM? Now 10000 RM? Nothing can grow at such an insane rate. Surely there is a dollar bubble, and it is going to crash.

What our German friend didn't realize was that his own currency was (hyper-)inflatory. USD were not overvaluated, they were stable. The current situation is similar, except instead of our currency being inflatory, the other one is being deflatory - this has the same effect on the exchange rate. It does not neccessarily mean there is tulip fever going on (although, to be honest, I can't disprove that).

Why aren't scarce resources like gold and oil going up the same way? I think gold made a low this week. Your theory doesn't align with what happened in Germany.
Exactly.

That German person could have bought a loaf of bread in the morning for RM1,000 then sold it to his neighbor in the afternoon for RM10,000. There are no citizens in Bitcoinland dealing with inflationary prices in virtual bread. Anything you can buy with Bitcoin is priced in fiat, denominated in BTC at whatever the spot rate of the moment happens to be.

EDIT:

> What our German friend didn't realize was that his own currency was (hyper-)inflatory.

He absolutely did. Just like Zimbabweans did more recently.

Gold is not growing in demand, and is not in the process of being "established". Almost everybody who wants gold already has enough, and people entering the gold market are approximately balanced by people selling gold. People who use gold (e.g. in industry) are supplying themselves at the same rate as they are using it. There is no reason it should continuously go up.

Of course my argument doesn't align with what happened in Germany, that is my point. What I am trying to say is what seems at first glance like an overvaluation or a bubble could actually be in-/deflation, or in other words: currency supply and demand.

If you assume bitcoin eventually becomes as big as a typical national currency, it is currently even undervalued.

For the theory to hold though, everything needs to go down in dollars. That's the point. If Bitcoin is the only thing going up there are no deflationary or inflationary forces at work.
You're right, there are not deflationary or inflationary forces at work. What is happening is simply supply and demand.

I'm thinking in a simple two-goods model here. "Currency" vs. "Products" - or "USD" vs "BTC". "Too much" supply of currency or "too little" supply of physical products drives the price up. In this model, that's the same as inflation, but you are right, as soon as there is a third good you can compare prices to that good and tell if it is inflation (all prices rise) or not (only the price of one good rises).

My point is not "it is inflation". Rather "it is inflation-deflation-supply-demand" (market reacting to changing supply and demand), as oppossed to "it is irrationality". If bitcoin is staying with us, and not banned etc., then demand for it will continue to grow faster than the supply will grow. I think the stable exchange rate of a fully established bitcoin will actually be at least a factor of ten higher than today.

It's technology and it's revolutionary. Both of those things tend to break all known patterns. Look at when corporations were invented or any other major tech revolution.

It could well be the first example of something going so high. It's unlikely but I wouldn't say it's 100% sure to crash.

(comment deleted)
(comment deleted)
I wonder how bitcoin will look in the aftermath of other technological novelties. Usually, these things start off as a niche novelty of an insular crowd, then savy people take part, then it goes mainstream and the influx of traffic forces regulation and more overhead due to demand. The drivers of the technology leave, then there's a slow burnout time where people try to maintain the appearence that the tech isn't dying.
You can't know that it's not a bubble. Maybe it is and crashes tomorrow. Maybe it is and crashes in a year. Maybe that crash brings it back to fundamentals but that price is still higher than it is today. If you could perfectly predict the future you'd be pretty much God.
Such price increases may be simply a bubble (possibly a temporary bubble as we've seen in the past), but they may also be a sudden price adjustment. One explanation for such a price adjustment is an increase in the number and wealth of its users. The user base has moved from crypto enthusiasts to wealthy individuals, and now it seems to institutional investors with bitcoin futures going ahead. Such a thing didn't exist before 2009, so we don't have much to go on as far as how valuable it'll be.

Cryptocurrency may not be pinned in value to anything tangible, but is useful and unique as an asset class for its ability to transfer ownership by virtual means. You can't store a bar of gold in your memory, and you can't send it through a computer. I believe its long term value is in how much people want money that has this property.

Long term, a threat to Bitcoin's value is competing altcoins. Bitcoin is arguably a bad cryptocurrency at a technical level (high transaction fees, slow transactions, bad for the environment, etc.), it's difficult to use for commerce, but seems to be winning out against the others for now, maybe just by first mover advantage and established network effect.

We only know bubbles in hindsight. No one can say for certain.
You can't know. For those who are saying it is obviously a bubble: If bitcoin "succeeds" and becomes widely used as money, then its value will grow a lot more, since there can only ever be 21 million bitcoins. It is inherently deflationary. One bitcoin is worth, roughly (and circularly) speaking, the total valuation of bitcoin divided by the number of coins. Another way of looking at it: imagine bitcoins become as widespread, and as desired, as USD. Now think about how many USD are in circulation compared to BTC.

I know it is not really that simple, and dollars have some special properties compared to bitcoin: you have to pay your US taxes in USD, for example. Also, it is easy to talk about "valuation" (in e.g. USD, EUR) if you view bitcoin as a commodity - but if it becomes a dominant currency it is not so simple. You can still reason about exchange rates, and my point is, the asymptotic exchange rate USD:BTC (assuming no dollar inflation) is not of the order of 10000:1, but much higher, maybe 1000000:1.

Bitcoin cannot become "widely used money". It is technically impossible. It is transaction rate limited to 6 txn/sec. Worldwide. Even band-aids such as increasing the block size won't help enough.

I'm not saying bitcoin prices will crash anytime soon. Simply that the dream of bitcoin replacing fiat currencies in daily commerce is just that: a dream.

It was once impossible to stream movies over the internet because modem technology was limited to 1200 bits per second. Even band-aids like doubling capacity wouldn't help enough.

Technically. Impossible.

Streaming movies was impossible over dial-up, and still is. But you can if you have different internet connections like DSL, Cable, Fiber, etc.

For the original bitcoin that is currently being priced at $13,000 it is indeed technically impossible to exceed 6 transactions per second. Sure you could make a new crypto protocol, but it's not "bitcoin".

The protocol (proof of work based system using wallets and addresses) doesn't define the implementation and the implementation defines limits. The protocol especially does not define the implementation of other layers built alongside it, so this is a straw man.

If your argument is "any improvement to the current system" "isn't bitcoin" then that's a no true scotsman sort of argument and I don't think it holds much water.

What will likely happen is that bitcoin will only be used to transfer large sums between exchanges. Day-to-day payments will be handled off-chain.

Another option is to use lighter alt-coins in parallel to bitcoin, again using bitcoin for long-term storage or larger transactions.

You are right that the current bitcoin as-is is not suited for many small everyday transactions, as envisioned by many. The transaction rate and fees are just too high. This is actually - besides governmental bans - the one thing I think could take bitcoin or cryptocurrency in general down.

So the US money supply is about $3.5 trillion. If you had 21 million BTC worth the same, your exchange rate would be about $170,000 / BTC.

Now I'll admit that I never expected BTC to break 5 figures, but even if your argument is "BTC should be worth as much as USD", 7 figures is still absurd.

Why would that be absurd? "Bitcoin" is an arbitrary unit, you can rescale it. For example, 1 satoshi = a hundred millionth BTC. In your example, the rate would be 1 kilosatoshi = $1.70. A nice and imaginable rate if it was a regular currency.

The thing is: you can either call current (and possibly future) bitcoin absurdly overvalued (with its real value close to zero), or the early-stage bitcoin absurdly undervalued (with its real value close to the "asymptotic" value). I personally prefer the latter, or better: acknowledge that the "value" is determined by the market, and the "real" value is dynamic in time.

The only question is: is the current exchange rate rational, or a craze? I believe, assuming nothing catastrophic happens (bitcoin outlawed, or it is abandoned because the network cannot handle the number of transactions anymore), the current rate is rational (close to the instantaneous "real" value), and will continue to rise (and approach the asymptotic "real" value) in future.

Full disclosure: I am not 100% convinced (and thus stupidly sold quite some BTC before the current hike), but I'm betting some coffee money that BTC - if it is staying with us - will go "to the moon".

"imagine bitcoins become as widespread, and as desired, as USD. Now think about how many USD are in circulation compared to BTC."

The US money supply (M1) is $3.5 trillion. The maximum Bitcoin supply is 21 million BTC. $3.5 trillion / 21 million BTC = $166,000 / BTC.

1 BTC being worth seven figures USD would mean that the "Bitcoin money supply" was 10x the "USD money supply".

(comment deleted)
> You can't know. For those who are saying it is obviously a bubble: If bitcoin "succeeds" and becomes widely used as money, then its value will grow a lot more, since there can only ever be 21 million bitcoins. It is inherently deflationary. One bitcoin is worth, roughly (and circularly) speaking, the total valuation of bitcoin divided by the number of coins. Another way of looking at it: imagine bitcoins become as widespread, and as desired, as USD. Now think about how many USD are in circulation compared to BTC.

This doesn't make sense. More precisely, your argument for why its value could grow more doesn't make sense. You're comparing incongruent units when you should be comparing per-unit value.

1 BTC is not an indivisible unit, it's just the most widely used unit. Similarly, we have $100 bills, $50 bills, etc despite $1 being the most widely used. We could also use pennies or nickels if the value of a single dollar far outstripped its ordinary utility for commerce.

The feedback loop you're suggesting would not occur because as the value of Bitcoin grew, each atomic slice of a Bitcoin would simply increase in utility as the new benchmark unit. With 100M satoshi per 1 Bitcoin, we're not in any danger of Bitcoin's valuation increasing just because there is a limited supply.

One of the biggest questions people have is how loans and debt can be handled in bitcoin. Since it's deflationary by nature, how will loans be easily recovered from the borrowers when it becomes harder to pay back the amount of BTC as the cost of BTC goes up?
When you look into bitfinexed articles regarding the Tether pump and dump and wash trading, you would be crazy to think the value of Bitcoin is sustainable.
The question should be, what should happen so that it stabilizes. If Bitcoin and cryptocurrency marker were to stabilize, the competition between different cryptocurrencies should mean the trade off between their different features will be directly reflected in the cost to their user. The possible features are: - Transaction time (Bitcoin is relatively bad here) - Transaction fee (Bitcoin is also one of the worst options here) - Smart contacts / other integration features (Bitcoin is going nowhere here) - Anonymity (Bitcoin isn't in the best spot here, might have the advantage of being the most commonly used BUT this is a matter of momentum only). - Resistance to attacks from large groups of people (Bitcoin's biggest advantage, but again only a result of it's initial momentum, which might not hold forever).

Now you can also notice that all these features are composite. If you really want to pay for anonymity, you can tumble though the most anonymous currency. If you need smart contacts, go through eth. Resistance to nation state attack is a noble ideological concept, but has low demand, and a cheaper but weaker coin will be dominant in most scenarios.

You will also know a market is stabilized when there no incentives to create competition without innovating. There's nothing stopping you from creating yet another crypto coin, which will have better transaction fee and will still be as decent as others.

Anyway you can clearly tell the market is clueless when the talk isn't about which coins gives you the best features for the most cheap price, but instead about which coin rises the most (which is usually exactly the opposite).

(comment deleted)