The Bitcoin Bubble and a Bad Hypothesis | Hacker News Reader

The Bitcoin Bubble and a Bad Hypothesis

nationalinterest.org

28 pointschrismealy56 comments

> willingness to hold the asset depends on the fact that it has value independent of that willingness. Tobacco can be smoked or chewed, gold can be used to fill teeth or make jewellery, and U.S. dollars can be used to meet obligations to the U.S. government.

Gold has very few industrial uses. If its price was its 'independent' value, it would cost a hell lot less. It's a value store because people believe is safe, even though it's much easier to steal a bar of gold than to fake bitcoins. Gold is also difficult to move around, or to make anonymous transactions with, or to split in small pieces etc. Gold is the old world's bitcoin, and, if anything its more flawed than bitcoins. As digital currencies prove their robustness and reliability, we will probably see a massive move from 'precious' metals to digital value stores.

This professor will be shaking his head in disbelief in a few years.

http://en.wikipedia.org/wiki/Gold_as_an_investment

Bitcoins may be hard to "fake", but not hard to steal. People are generally more familiar (currently) with physical security then digital security.

Yes and no: it is just a question of habit. You will no more need some physical coffer. You will just need to remember a big password (or just a small coffer where you put a unique printing of your password). The protection of your property will be cheaper.

It would be interesting to see statistics on this. Does digital theft happen more often than physical? In the end, you can always encapsulate a wallet in a crypto physical device.

Calling the Bitcoin Bubble "final refutation of the efficient-markets hypothesis" is a stretch. MtGox is opaque and illiquid for something that looks like a stock market, and the market is tiny and centered around the very long-shot possibility of distant future value.

If you can't arbitrage a a market, EMH puts no bounds on how wrong it can be, and mtgox is very difficult to arbitrage, so the price wanders around chaotically between prices extreme enough to actually make risk-adjusted profit off them.

The situation for options/futures is even worse than at mtgox, and a broad acceptance of bucket shop behaviors based on mtgox market feeds can't be helping. From top to bottom, this is not a market oriented towards the efficient discovery of prices.

I think the main point about the EMH and Bitcoin is that a market can be quite efficient without anyone knowing that it is efficient or why this is so...

Bitcoin does have an intrinsic value: the ability to buy things online privately. People have a need to buy things online without their credit card company using that information to target marketing or update their risk profile.

People continue to use cash in the physical world, even though it's less convenient than cards, because it is private and the merchant gets 100% of the money. Bitcoin has the same virtues online.

FWIW, you can't actually pay your taxes with cash.

You're misinterpreting the definition of intrinsic value. The ability to buy things online privately is something entirely extrinsic to BTC. Even people who say that the US dollar has intrinsic value (aside from the paper) are mistaken - value derived by fiat only gives the dollar legal value.

You most certainly can pay taxes with cash. But that's not the point. The point is that you have to pay your taxes in US dollars.

To be fair, the article recognizes intrinsic value in the dollar, but not in bitcoin, applying double standards. Btc 's divisibility, the fact that it s hard to counterfeit, and anonymity are intrinsic values although the whole concept of intrinsic value is vague, because in the end all values are market-determined

That doesnt make sense. The dollar has extrinsic value too, which comes from the legal requirement of governments to accept it as payment. Unless you feed on paper, it doesn't paper money has less intrinsic value than digital, the latter is more divisible and easily exchangeable worldwide.

http://en.wikipedia.org/wiki/Intrinsic_value_(numismatics)

You are correct that intrinsic value is very vague. That wikipedia article, for example, is incorrect if you apply a strict definition of intrinsic value. Since all currency value is essentially subjective, all fiat currencies / digital currencies are devoid of intrinsic value.

The value from satisfying legal requirements is extrinsic.

And the value of the ability to fill your mouth with gold is completely extrinsic to gold.

> Bitcoins will attain their true value of zero sooner or later, but it is impossible to say when... Bitcoins are the most demonstrably valueless financial asset ever created

It looks like John Quiggin is basing his views upon his extensive prior research[1] into Bitcoin, and he is not simply jumping onto the bandwagon like so many other economists.

[1] http://johnquiggin.com/?s=bitcoin

Since in the OP he fails to explain any reasons why anyone might find Bitcoin useful or valuable in the first place, one could doubt the quality of his 'extensive prior research'.

No, non-monetary value of Bitcoin is in ability to store a proof of anything in its blockchain without trusting any third party. E.g. timestamping documents, registering contracts, registering intellectual property etc. See more here: http://blog.oleganza.com/post/42262765318/direct-use-value-o...

PS. "intrinsic" value does not exist in anything. Value is something that people put in things. Nature itself has no meaning, values or purpose. It's we, the actors who make the choices and valuations.

Yes, you are absolutely right with Bitcoin being a means of proving a transaction. But the journalists aren't going to pick up on that distinction soon.

Even aside from its utility in commercial exchanges, bitcoin has an arguably minor but often-overlooked "intrinsic"* value. A bitcoin is nearly irrefutable proof that a certain amount of difficult computation has been done by (or on behalf of) its owner. That has some inherent value, in the vein of Hashcash, for potentially rendering spam or DDOS attacks unprofitable. But unlike Hashcash, the work can be done long before hand, and there is a decentralized network preventing double spending.

* I put "intrinsic" in quotes to broadcast that I am aware of the philosophical and semantic difficulties the word introduces into a conversation. Do apples have intrinsic value because people can eat them? What if I'm allergic to apples? What if the human race goes extinct?

http://en.wikipedia.org/wiki/Subjective_theory_of_value

> advances the idea that the value of a good is not determined by any inherent property of the good, nor by the amount of labor required to produce the good

This is what most economists agree upon.

Indeed, and yet, you see so many bitcoin articles in major publications talking about bitcoin's lack of intrinsic value.

Which is completely true: bitcoin has no intrinsic value (as do all fiat currencies).

I have yet to hear a definition of "intrinsic" that doesn't require knowledge of the future or include either everything or nothing.

Today, Bitcoin has no intrinsic value.

you can pay your taxes with cash, but you have to do it at a physical IRS office.

http://www.irs.gov/Businesses/Small-Businesses-&-Self-Em...

You're right. I was thinking of Canada, where they eliminated the option several years ago.

> Bitcoins will attain their true value of zero sooner or later, but it is impossible to say when.

I predict it'll happen the second everything is made legal everywhere in the world.

When this happens we can still use it until international wire transfers are free :)

So i'm not the only one that noticed the writer forgot to mention what bitcoin is mainly used for :P

It is quite incorrect to say that "Historically, money arose from, and in conjunction with, this power." (the power of taxation by states). He gives no evidence for this claim except to reference a "controversial" book - perhaps controversial because of its wildly inaccurate thesis?

Unless the contention is that humans traded solely by barter before states came along to tax them, mediums of exchange (money) obviously predate the statist invention of fiat money. In fact fiat money is a relatively recent phenomenon, as money was historically backed by precious metals long before states invented unbacked fiat money and central (statist) banking.

Which doesn't in any way invalidate his argument that Bitcoin is a bubble. It's totally irrelevant to the Bitcoin issue. It's just wrong by itself.

Another thing that's just plain wrong is the attempt to refute the efficient markets hypothesis (EMH). The only reason Bitcoin actually does have a value above 0 at the moment is precisely because statist meddling in the market has created a huge demand for an alternative to fiat money - such a huge demand in fact that people are willing to try almost anything, including Bitcoin, or even riskier methods with a high failure rate (such as J. Orlin Grabbe's early anonymous banking system DMT, or unaudited gold-backed e-currencies that may actually be backed by nothing at all, among many other examples).

If we had uncrippled free-market gold-backed electronic currencies (such as e-gold was before it was attacked and destroyed by the feds) there would be zero demand for bitcoin as the existing currencies would have all the benefits of Bitcoin, while also being backed by a commodity with a base value (not 'intrinsic value').

So, it _might_ be a refutation of EMH if Bitcoin were to be successful long-term in an actual free market. As nowhere on this planet do we find anything even remotely resembling an actual free market, his argument against EMH is as much of a straw man as all the other examples that pinkos trot out as examples of "market failure" while pretending that there is a free market somewhere, which there isn't. All such examples are really examples of statist intervention failure, not market failure, as the market we are talking about is highly crippled and not remotely free.

This is just fetishism.

All currencies are fiat currencies, and all currencies are creatures of the state. If a currency is set at a fixed exchange rate to a precious metal, it's because a "statist fiat" made it so.

Not at all. Gold by itself is a currency / medium of exchange. It doesn't need any "statist fiat" to make it so. The fact that states (or other entities) might issue redeemable paper certificates or receipts (or dollar bills) as a convenient stand-in for gold deposited with them is entirely irrelevant. Gold is already money before that and without that.

Gold is only "money" inasmuch as it is accepted as currency, to pay debts, etc. If cowrie shells were as acceptable then they'd be just as legitimate money as gold.

Or for a better example, cigarettes as currency in prison.

Sure, but this has nothing to do with the point that gold is not money by statist fiat.

Gold's real scarcity and base value are what make it suitable for use as money and therefore universally accepted as money for thousands of years.

Cowrie shells might also make an acceptable non-fiat currency if cowrie shells were really scarce (they're not, because one can presumably breed cowries and generate them at will - something that was perhaps not so practical at the time when they actually were used as currency) or had a significant base value (again, they don't though they once did, but purely for ornamental uses). Similarly cigarettes are a currency in prison without any fiat being involved because they're scarce in that context and they have a base value (based on their use for smoking).

Bitcoin has neither real scarcity nor any basis for a lasting base value. It has illusory scarcity, and a transient base value based on the fact that in the current historical context, it enjoys network effects and near zero competition for certain use cases (illegal transactions where fiat currencies are unsuitable).

There's nothing sacred about gold either. If efficient nuclear transmutation of lead into gold were possible, gold would no longer be suitable as a currency. We'd have to switch to Latinum.

> According to the efficient-markets hypothesis (EMH), which still dominates the analysis of financial markets, this should be impossible. The EMH states that the market value of an asset is equal to the best available estimate of the value of the services or income flows it will generate. In the case of a company stock, this is the discounted value of future earnings. Since Bitcoins do not generate any actual earnings, they must appreciate in value to ensure that people are willing to hold them. But an endless appreciation, with no flow of earnings or liquidation value, is precisely the kind of bubble the EMH says can’t happen.

No. No no no no no.

EMH doesn't say that everything must be priced according to the "best available estimate of the value of the services or income flows it will generate." For example, if I pay $20,000 for a car, that doesn't mean that the present value of the extra income I'll earn thanks to that car is $20,000. It might mean the car makes my life more pleasant and convenient. Dollars are valuable for exactly the same reason--since everyone I know will accept dollars instead of whatever it is they'd like to buy with dollars, having dollars makes my life extremely convenient. That, not the fact that I need them to pay taxes, is what makes dollars so valuable.

The issue here is that there are network effects. Just as dollars are more valuable to me because they're used and accepted by everyone I know, or a fax machine is only valuable to me if I know someone who can receive my faxes, a currency is only valuable when someone else is willing to accept payment in it. This means that, as a currency is adopted by higher and higher proportions of the population (or as the perceived odds of that rise), its value also rises.

Bitcoin definitely looks like a bubble to me. It's a small market, it's dominated by people with strong philosophical positions that predispose them towards BTC. And shorting is difficult. But the reasoning in the article is absurd.

The EMH applies to market-traded securities. Your car is not a marketable security.

That's probably a better way to put it. I guess the point is that EMH doesn't apply to BTC for some of the same reasons it doesn't apply to cars--both cars and currencies get much of their value from utility that's unrelated to the income flows they generate.

The EMH applies when there is a robust market, as there is in bitcoin. How assets get their value to each participant or what valuation methods are used is irrelevant -- the point of the EMH is that the market price is correct because of the market. I don't believe the EMH is accurate, but it's worth understanding exactly what it is if you're going to say it is wrong.

The EMH is routinely applied to currency trading, as well as asset trading (i.e. gold/silver/natural gas). It doesn't matter whether you view BTC as a medium of exchange or a store of value -- the EMH is reasonably applied against BTC because there is a robust market.

It looks a bit silly to accuse me of not understanding what EMH is while simultaneously changing your working definition of its applications from "market-traded securities" to "liquid markets." I accepted your scope for EMH and showed that the author's argument was still incoherent--currencies are not securities[1]. But if you'd rather go back to a scope for EMH that applies to BTC, that's fine. The author claimed that:

1. EMH applies to BTC.

2. This means that the price of a BTC should be the value of its underlying asset.

3. The value of BTC's underlying asset (nothing) is 0.

4. Therefore, EMH is wrong.

And as my original post argued, 2 is wrong. In cases where an asset has value external to its backing, EMH says nothing of the sort. A corrected version of the author's logic:

1. EMH applies to BTC.

2. This means that the price of a BTC should capture the present value of a BTC from several sources--the value of its underlying asset, its utility as a medium of exchange, and the any expected appreciation in the future.

3. The value of BTC's underlying asset (nothing) is 0.

4. Therefore, either BTC has positive utility as a medium exchange, BTC is expected to appreciate in the future, or both. Or EMH is wrong.

Either we apply EMH to market-traded securities, in which case it doesn't apply to BTC, or we apply EMH to anything with a liquid market, in which case it doesn't say that BTC's true value is the value of its underlying asset. Either way, the argument falls flat as a refutation of EMH.

[1] http://en.wikipedia.org/wiki/Security_(finance)#cite_note-1

I'm not saying your conclusions are wrong. I'm only saying that your application of the emh is off by a few degrees.

The emh applies to liquid markets. The market for cars is illiquid and therefore the emh doesn't apply. There is no contradiction in these, I never said the emh only applies to securities. I'm sorry if how I said it was unclear or confusing.

There is a robust market for BTC, therefore the EMH is reasonably applied. This doesn't mean I'm agreeing with the author (I agree that the premise that BTC has a knowable zero value is tenuous). My argument is very limited in scope.

http://en.wikipedia.org/wiki/Subjective_theory_of_value

We are arguing about the wrong two questions: 1) What is the right value of 1 Btc? $0 or $100, or $1000? 2) Can it replace government issued currencies?

The right question to ask is 3) What is Bitcoin good for?

It's not currently good at storing value over the long term. Short term it worked most of the time.

And that's been the pattern in the bitcoin economy: Transfer globally, or in transactions where other payment options are prohibitively expensive, and exchange bitcoin for your local currency by local and transparent means.

The fact that he calls the computations "useless" in his second paragraph proves he doesn't understand the concept. Miners are being rewarded for double-checking the validity of recent transactions... If he said the same of Visa's or American Express's computer computations he'd never get published!

Bitcoin intrinsic value:

It can be transferred (spent) electronically over a decentralized network without double-spending risk, chargeback risk, inflation risk and confiscation/bailout/haircut risk.

How can anyone say this has no intrinsic value? You can say that you don't care about this properties, like I don't care that gold is shiny, but that doesn't mean it has no value.

I think many analysts confuse intrinsic value with being tangible.

First, as has already been pointed out, nothing has intrinsic value. Things only have value inasmuch as somebody values them. Value is a relationship between the subject and the object, not a property of the object.

Gold has a 'base value' that arises because of its demand for non-monetary uses, such as its industrial uses or use in jewellery. This base value is what makes it suitable as a medium of exchange as the base value sets gold's initial non-zero exchange rate vis-a-vis other goods. Once gold (or any other commodity) starts being used as a medium of exchange, it generates further demand AS a medium of exchange, and this will increase its price in relation to other goods, giving it a currency value higher than the base value.

It's like mathematical induction - where the general case is predicated on the base case. If there is no base case (or base value) there is no foundation for the general case (or for a currency value).

Bitcoin has no base value. Yes it "can be transferred (spent) electronically over a decentralized network without double-spending risk, chargeback risk, inflation risk and confiscation/bailout/haircut risk". But so can Bitcoins based on an alternate blockchain. What makes the 'official' blockchain Bitcoins have any higher base value than the worthless bitcoins on an alternate blockchain?

In fact, I expect there will be alternate blockchains in the near future, and that is what will burst the bubble as people start to realise that the number of bitcoins may be limited, but the number of bitcoin-like currencies is unlimited, so really, the limit on the number of possible bitcoins is entirely illusory and meaningless.

A free-market, decentralized fiat money is still more useful for many purposes than statist fiat money, and for this reason Bitcoin will enjoy some popularity until a proper anonymous, gold-backed free market e-currency comes along. When that happens, there will no longer be any use case for Bitcoin and its currency value will collapse to its base value of 0.

Two comments:

1. Without the decentralized network that you mention, a bitcoin has no value. But the decentralized network requires some level of acceptance for anyone to run it. So bitcoin's value depends on its acceptance, in a way that gold does not.

2. Bitcoin certainly incurs inflation risk - the large recent drop in the purchasing power of a bitcoin is inflation, after all. Perhaps you meant "without risk of manipulation by the currency issuer," but currencies may move for many other reasons.

Seems like the author never really understood bitcoins.

Bitcoin is not a bubble. It's a way to move money efficiently on internet.

The speculation around bitcoin was a bubble. Bitcoins will still remain used and useful as long as people use them.

"Bitcoin is not a bubble. It's a way to move money efficiently on internet."

The two are by no means mutually exclusive.

This is exactly what I wrote.

IMHO bitcoin provides immense value once it is truly anonymous. If anyone can anonymously gamble, buy and sell goods, and trade against bitcoin for profit then it's a perfect currency for illegal activity, which in itself represents a massive economy outside the walls of gov't.

Proving once again that the economics departments are about as useful as the philosophy departments.

(No offence to those few good academic economists and philosophers.)

I don't know that you would be insulting actual economists. You learn pretty quickly that EMH is an abstract model and should be treated as such. It is unfortunate that such caveats rarely make it across campuses into business/finance departments. The article itself clearly falls victim to this and commits the further sin of casually dismissing the transactional uses of bitcoin.

EMH has issues and bitcoin may fail, but it is unlikely to be for the reasons the author cites.

Interesting analysis. More discussion at Quiggin's blog here: http://crookedtimber.org/2013/04/17/bitcoin-the-perfect-bubb...

"But John, last week I had x Bitcoin. Today I have x Bitcoin. So value is unchanged. Constant at x Bitcoin. Dollar is volatile though. #ThinkBTC"

That comment is a reference to another post on the front page of Crooked Timber right now: http://crookedtimber.org/2013/04/17/annals-of-the-gold-stand...

Not sure you'll see this, but I am curious, do you ever see a link that has no discussion and that you think deserves a discussion, or you think has the potential for a discussion you would like to read?

If so, do you ever attempt to "rescue" them by leaving a somewhat neutral comment of some value?

... which is why we usually compare currencies over time in terms of baskets of goods.

That's an optimist's way of looking at it. Look at how strong the USD is against BTC. Things must be great in America!

Things are great in America, compared to most of the world, actually ;-)

Reply on news.ycombinator.com