|
davidgdg
|
 |
November 07, 2014, 09:05:19 AM |
|
"SINCE its inception, the market for bitcoin has experienced one of the most remarkable roller coaster rides of all time. Originally worth a fraction of one cent in 2009, the price of bitcoin rose to peak at not far short of $1,200 in December 2013. Since then, it has fallen erratically and is currently about $328. Over its short history, however, the price of bitcoin has repeatedly risen and fallen and then risen again."
Correct.
"So is this recent fall just another one of its market corrections, before it resumes its spectacular rise? Or does it suggest that the bitcoin bubble will continue to deflate or possibly even burst?"
A reasonable question.
"To answer this question, we need to look at the fundamentals."
I agree.
"A glaring problem then becomes apparent. To work as intended, the system requires competition on the part of the bitcoin “miners”, who validate transactions blocks in their search for newly mined bitcoins. Miners compete to validate the most recent transactions block – one block is validated approximately every 10 minutes – and the first one to do so is rewarded with newly created bitcoins."
It is unclear what "requires competition" means here. Conventionally, "competition" refers to suppliers competing to attract customers by offering improved services or reduced prices. However, mining is not a market in this sense (there is a debate as to how transaction fees might be marketised, but that is an entirely different issue). The protocol requires only that work be done by miners in order to reach an honest consensus on the state of the network. This process does not require competition. Presumably what KD is referring to here is the risk of attacks in the case of a single node controlling > 50% of hashing power.
"However, the mining industry is characterised by large economies of scale – these are so large, in fact, that the industry is a natural monopoly, in which it is economically more efficient to have one producer than many."
It is unclear why this should be the case. Efficient hashing is a function of access to cheap energy, skilled labour and advanced capital equipment. Indirect costs may include regulatory burdens and the cost of network access. None of these features are unique to bitcoin mining and it is entirely obscure as to why they entail the consequence that a single producer would tend to dominate. KD does not explain any of this. One might give the example of access to cheap hydro-power in Iceland. Why should this entail the existence of a single dominant hashing firm?
"The problem is that these economies of scale are inconsistent with long-run competition. This implies that the bitcoin system is not sustainable and must therefore collapse."
This is obscure. The existence of a single or small number of firms does not usually entail the collapse of a particular industry.
"Indeed, these centralising tendencies are already playing out in the bitcoin mining industry. Mining pools are now so big that the original atomistic competition has given way to oligopoly, and there is concern that these mining pools are big enough to threaten the system by subverting the transactions validation process for their own ends: for example, by mounting some kind of double-spend attack."
The existence of mining pools has nothing to do with economies of scale. It is a function of reward variance and risk aversion. Pools need not perform any hashing of their own. Thus the alleged economies of scale which are said to characterise the hashing industry, are irrelevant. Perhaps what KD means to say is that the mining pool industry (not to be confused with physical hashing by mining entities) is itself characterised by economies of scale. In other words, the larger the pool the lower its internal costs. But it is unclear why this should be so and KD does not explain. What is the characteristic of pooling that entails the proposition that internal costs are inversely proportional to pool size, without limit? This proposition is central to KD's argument but he does not address it at all. It is difficult to rebut an argument when one does not know what it is.
"Up to now, most of the big players have managed to maintain confidence in the system by exercising self-restraint thanks to a shared belief in bitcoin idealism – in effect, a commitment to the bitcoin community as a whole. However, in recent months, one big pool, GHash.IO, has openly rejected that idealism and now poses a major threat to the system."
The extent to which GHash posed (or poses) a threat of attack is a matter of conjecture and has been addressed elsewhere. The more interesting and relevant question to KD's thesis is why and how Ghash succeeded in attaining almost 50% of hashing power. Was this a simple function of lower internal costs with size? And if so, what are the features of the poooling industry that entail this state of affairs? Or was the success of GHash a transient state of affairs reflecting an aggressive and innovative business model, not yet matched by any competitor? If the latter, then in the absence of barriers to entry, one would expect competition within the pooling industry to make inroads into GHash's market share, as with any other industry. This seems to be happening at present. At all events, this critical question remains entirely unaddressed by KD.
"But even without GHash, the situation in the mining industry is unstable, and the underlying economics – the tendency to monopoly – must eventually win out: even bitcoin cannot defy the laws of economics forever."
See comments above. KD does not identify the "laws of economics" which entail the proposition that mutual pooling to reduce variance is a natural monopoly or explain why this should be so.
"The bitcoin system is thus already reduced to the point where it is relying on trust in the dominant mining pools not to abuse their power."
This does not appear to be the case at present.
"Distributed trust – trust in the network itself, rather than in any individual players – is living on borrowed time."
This states a conclusion rather than an argument.
"However, distributed trust is the core of bitcoin’s value proposition."
I agreed that this has conventionally been regarded an important element of the bitcoin system but its absence would not necessarily be fatal.
"Once that goes, one can then see the other elements of its value proposition – the absence of a single point of failure, transactions anonymity, even the integrity of the bitcoin protocol itself – falling one by one like dominoes. There will then be no point in using bitcoin anymore: if you have to rely on a big mining pool to keep its promises, then you may as well use PayPal instead."
I'd suggest this proposition is false. The bitcoin ecoystem offers other features beyond distributed consensus which PayPal cannot duplicate. Paypal is a relatively high fee, tightly controlled value transfer firm. The blockchain offers many features beyond that. The partial loss of distributed consensus (were it to occur) would not invalidate the existence and value of those other features. It is perfectly plausible to conceive of a semi-centralised feature-rich blockchain dominated by a benign single pool competing very effectively with Paypal and other legacy systems. That said, until KD explains why such a dominant entity might be expected to emerge, the issue is hypothetical.
"In the meantime, there is nothing within the bitcoin system to credibly shore up confidence."
This is obscure. Confidence in what?
"The sticking plaster solutions that have worked up to now cannot work indefinitely."
This is a further restatement of KD's conclusion rather than an argument.
"At the same time, unlike gold or tulips, bitcoins have no alternative use."
The blockchain has uses which are unrelated to its monetary or even its asset representation function. Timestamping of hashed digital data is an example.
"So when the price collapses, there is no “natural” price to fall to north of zero. And so the whole bitcoin system eventually becomes a house of cards, and anything – a scandal, a government attack, whatever – could trigger a loss of confidence leading to a run that brings it all down."
The same is true of fiat. This could just as well be a description of the process of hyperinflationary fiat collapse, which I would suggest is considerably more likely than a complete loss of confidence in bitcoin as a result of domination of hashing power by a single benignly disposed pool.
"In fact, a number of the “alt” cryptocurrencies – including Auroracoin, Terracoin, Freicoin, Mooncoin, Scotcoin, BBQCoin and several others – have already experienced massive price collapses. There is no reason why bitcoin should be exempt from the same fate."
KD does not analyse the causes for the price collapses of these alt currencies or make any attempt to analyse whether those causes have any relevance to bitcoin. The statement is analogous to the proposition that since the price of company stocks A to D has experienced massive collapse, the same might be expected to be true of the stock of company E. Without knowing why the price of stocks A to D collapse, the statement is an empty one.
"For any investor, the rational decision is to sell before the roof falls in. If enough think the same way – and why shouldn’t they? – then their expectations will become self-fulfilling. There will be a stampede for the exit, the price of bitcoin will drop to its intrinsic value – zero – and the system will collapse."
True if but only if (1) economies of scale within the mining pool industry increase with increased size, without limit (a proposition unexplained by KD) (2) the competitive advantage of bitcoin springs entirely from distributed consensus (a proposition which appears to be false).
"The only question is when."
I suggest there are quite a number of other questions that KD needs to address first, before posing this one.
|
