MetcalfeвЂ™s Law, which measures the value of a network, can calculate a cryptocurrencyвЂ™s valueвЂ”and predict when to get out.
- by Emerging Technology from the arXiv
- March 29, 2018
Philosophers, economists, and theorists have various ways to judge how money should be valued. Some have said that its worth lies te a high cost of production. Others see it spil simply a form of credit that permits the transfer of resources, which is why it can take the form of chunks of paper or even digital records.
Recommended for You
Then there is the idea that a currency is worth whatever somebody is willing to pay for it given the limited supply. This explains the extreme valuations sometimes seen for the cryptocurrency Bitcoin.
All thesis approaches run into trouble of one form or another. There is certainly a high cost of production ter the cryptographic вЂњproof of workвЂќ required to create, or mine, bitcoins. But their value has little relation to this cost. By the end of 2018, a single Bitcoin wasgoed worth almost $20,000, and the cryptocurrency market spil a entire had a value of $830 billion. Just a few weeks straks, the market had collapsed to $280 billion.
So how much is a Bitcoin truly worth?
Today wij get an reaction of sorts, thanks to the work of Spencer Wheatley at ETH Zurich te Switzerland and a few colleagues, who say the key measure of value for cryptocurrencies is the network of people who use them. WhatвЂ™s more, they say, once Bitcoin is valued ter this way it becomes possible to see when it is overvalued and perhaps even to spot the telltale signs that a market crash is imminent.
The value of a network is famously accredited to Bob Metcalfe, the inventor of Ethernet and founder of the rekentuig networking company 3Com. MetcalfeвЂ™s Law states that a networkвЂ™s value is proportional to the square of the number of its users.
ItвЂ™s straightforward to calculate a value for Bitcoin based on the number of active users. Wheatley and co getraind the gegevens to a generalized MetcalfeвЂ™s Law that permits them to tweak the parameters, arriving at an exponent of 1.Sixty nine rather than MetcalfeвЂ™s original square of the number of users (i.e., an exponent of Two).
This makes sense. The original law is based on the idea that the value of a network grows te proportion with the number of all possible connections. Te other words, it assumes that all knots can connect with each other.
вЂњThis does not seem realistic,вЂќ say Wheatley and co. Their finding is that each user is on average linked to N Two/Three other users. вЂњFor example, for N = 1 million, a typical user is then connected to вЂonlyвЂ™ Ten,000 other users, a more realistic figure,вЂќ they say.
With thesis parameters, the generalized MetcalfeвЂ™s Law more accurately reflects the way BitcoinвЂ™s value has enlargened with the number of users.
It also exposes when Bitcoin has bot overvalued. Wheatley and co point to four occasions when Bitcoin has become overvalued and then crashed, te other words, when the bubble has burst.
Thesis events have bot well documented. The very first big crash occurred ter 2011 when Mt. Gox, a major Bitcoin exchange te Tokyo, wasgoed hacked, presaging an 88 procent druppel te the cryptocurrencyвЂ™s value overheen the next three months.
A crash te 2012 wasgoed preceded by the discovery of a Ponzi fraud involving Bitcoin. Another crash occurred te 2013 when high trading volumes shocked Mt. Gox, causing it to collapse, the value of Bitcoin then dropped by 50 procent te two days.
The most latest collapse, at the end of 2018, occurred after South Korean regulators threatened to shut down cryptocurrency exchanges.
To explore thesis collapses, Wheatley and co use a monster developed by Didier Sornette, who is the professor of entrepreneurial risks at ETH Zurich and one of this paperвЂ™s authors. Sornette has long suggested that it is possible to predict the collapse of speculative bubbles using certain characteristics of the markets. Indeed, readers of this blog will be familiar with his ideas.
SornetteвЂ™s treatment has two components. Very first, he looks for markets that are growing at a super-exponential rateвЂ”in other words, markets where the growth rate itself is growing.
That can toebijten for brief periods of time because of factors such spil herding behavior. But it is not sustainable without an infinite number of people. For this reason, a crash, or correction, is unpreventable.
This much is uncontroversial. But Sornette goes on to say that the timing of the crash is predictable. ThatвЂ™s because the unsustainable growth rates leads to hefty volatility. And this makes the market increasingly unstable, to the point that almost any puny disturbance can trigger a crash.
So te the Bitcoin crashes listed above, the triggering events are insignificant. According to Sornette, the market wasgoed already te a critical phase, and if thesis events hadnвЂ™t occurred, some other event would have triggered a crash instead.
The situation is analogous to a forest fire. If the forest is dry enough to burn, almost any spark can trigger a blaze. And the size of the resulting fire is unrelated to the size of the spark that commenced it. Instead, it is the network of connections inbetween the trees that permits the fire to spread.
The controversy overheen SornetteвЂ™s work is how accurately he can make thesis predictions. Clearly, a prediction that Bitcoin is about to crash ter the next few hours or days is much more powerful than a prediction that it will crash te the coming months or years.
Nevertheless, the researchers say it permits them to predict market crashes using gegevens from the past and so should permit them to spot similar imminent crashes te the future. They waterput it, rather confusingly, like this: вЂњ[Our] specimen is shown to provide an ex-ante warning of market instabilities, quantifying a high crash hazard and probabilistic bracket of the crash time consistent with the actual corrections, albeit, spil always, the precise time and trigger (which straw cracks the camelвЂ™s back) being exogenous and unpredictable.вЂќ
According to the generalized MetcalfeвЂ™s Law, Bitcoin is significantly overvalued, even after the crash at the end of 2018. вЂњOur Metcalfe-based analysis indicates current support levels for the bitcoin market te the range of 22вЂ“44 billion USD, at least four times less than the current level,вЂќ they say.
And that means there is uncertain weather ahead, at best. Wheatley and co compare the current Bitcoin market conditions to those following the collapse of the Mt. Gox trading system. вЂњThe current market resembles that of early 2014, which wasgoed followed by a year of sideways and downward movement,вЂќ they say.
That sends a not-altogether-unexpected message to Bitcoin miners, speculators, investors, and potential regulators: Beware!