Our Privacy Statment & Cookie Policy

All LSEG websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.

The Financial & Risk business of Thomson Reuters is now Refinitiv

All names and marks owned by Thomson Reuters, including "Thomson", "Reuters" and the Kinesis logo are used under license from Thomson Reuters and its affiliated companies.

June 1, 2017

The Market Sentimentalist: Bitcoin Bubble Watch

by Amareos.

Since January the price of Bitcoin has surged more than 90%, taking it over the symbolically but not economically meaningful USD2,000 mark (at the time of writing it stands at USD2,400). As so often happens when Bitcoin’s price breaks through a big round number, and especially when it coincides with a rapid acceleration in price momentum, it provokes a slew of media coverage decrying it as an intrinsically worthless[1] financial instrument more akin to a Ponzi, or pyramid, scheme. The near parabolic rise in Bitcoin’s price recently – see exhibit below – certainly does appear consistent with bubble dynamics.

Exhibit 1: Bitcoin Price – USD

Source: www.quandl.com

While accepting there will be periodic (meaningful) pullbacks in Bitcoin’s price, we disagree with the bubble conclusion for several reasons.

First off, let us dismiss the notion that Bitcoin is nothing more than a technologically sophisticated Ponzi scheme that implies, almost by definition, a bubble. This perception mainly arises because positive returns from holding it are generated via future price appreciation[2]. Hence, the naysayers argue Bitcoin relies on the “greater fool theory”.

Our rationale for rejecting the Ponzi analogy is that unlike such schemes Bitcoin does have useful attributes. Even though it doesn’t attract a nominal interest rate set at the discretion of the central issuing authority (ie. government-backed central bank) like fiat money, and therefore generates a return via future price changes, it has many of the prerequisites of money. It is no more a Ponzi scheme than all the other fiat monies in existence whose usage is simply reflective of their common acceptance (aided, of course, by official compulsion as they are legally accepted means of extinguishing liabilities, especially tax[3]).

Bitcoin is portable, divisible[4] and a medium-of-exchange. Moreover, its acceptability – another prerequisite for money – is increasingly recognized by governments as evidenced by the recent passage of legislation by Japan’s Diet which brings virtual currency exchange operators under the purview of the Financial Services Agency. Such legislative changes have boosted demand from Japan, which many consider the key driver behind Bitcoin’s recent price surge.

Critics of Bitcoin argue that its medium-of-exchange functionality is problematic for two key reasons. The first relates to Bitcoin’s structure which means there is limited transaction volume capacity – a maximum of 7 transactions per second which compares with thousands of transactions conducted globally over the same time interval[5]. We are no experts on the issue of scalability but looking at Ethereum – another virtual currency whose price is soaring in value – some of the technical impediments to increasing transaction bandwidth may not be insurmountable, but at present it is clearly a constraining factor.

To understand the second issue with Bitcoin’s usefulness as a medium-of-exchange we have to engage in a little intellectual detour and examine another of its key money attributes, one that is a direct result of one most misinterpreted aspects of its design: only 21 million Bitcoins will be produced, ever![6]

The (eventually) fixed supply feature has been described by some as a fatal design flaw[7] because, in addition to making it an inherently deflationary monetary system, Bitcoin’s attractiveness as store-of-value is perceived as undermining its medium-of-exchange function.

Of these, the deflationary aspect is the least concerning to us. For this to be valid Bitcoin would have to supplant all existing fiat monies. This is impossible to image. Governments would not tolerate such an outcome, not to mention the obvious profit motives for creating additional virtual currencies as is already occurring[8].

In relation to its usefulness as a medium-of-exchange, this critique is predicated on Gresham’s law that “bad money drives out good”. The underlying principle is given the choice to transact in a currency that is expected to depreciate or a currency that is expected to appreciate it is rational to use the former for transaction purposes and hoard the latter.

As Bitcoin’s total supply is designed to be fixed at 21 million coins, it is reasonable to expect, assuming unchanged private-sector holding preferences (a biggie we know), that Bitcoins’ purchasing power will be maintained when aggregate prices are rising (ie. positive inflation rates). Compared with fiat money, whose guardians – central banks – are mandated to ensure that the purchasing power of paper money falls[9] (in its electronic form one must also incorporate the nominal interest rate although this has been fairly negligible in many developed countries over recent years), it is not hard to see why this is an attractive design feature.

This is a perfectly valid argument. That said, the severity of the impact crucially depends upon Bitcoin’s relative attractiveness as a store-of-value versus its medium-of-exchange use. This naturally raises the question of what is the appropriate fundamental valuation for Bitcoin?

With no clear precedent this is not a straightforward exercise. One attempt by an unnamed financial analyst was outlined in a recent FT alphaville blog post[10]. The basis of the valuation method was Bitcoin’s use in the black market economy[11]. Using estimates of money laundering weighted up by nominal GDP and benchmarking Bitcoin’s price against this “illicit activity metric” a simple valuation measure was created. Over the period 2014 to the present day, this so-called Bitval ratio averaged 69. On the back of the recent price surge, this ratio has risen to 164 prompting the author to conclude that Bitcoin is overvalued by almost 240%.

Buuuubbbbbbbbllleee!

The author acknowledges that this is only a first attempt at generating a valuation for Bitcoin, however the flaws with this approach are pretty clear.  First, data on the black economy are very dubious for the obvious reason that illegal activities tend to be shielded from governments. Hence, is extremely difficult for national statistics agencies to collate accurate data.

Second, and in our opinion more problematic, is the assumption that Bitcoin transactions are almost exclusively related to black market activity. This may have been true historically but, as mentioned above, there is increasing acceptance of Bitcoin and other virtual currencies by the wider general public and even governments. Hence, the fundamental premise of this approach is flawed making the resultant valuation meaningless.

So, what about alternative ways to generate a fundamental valuation for Bitcoin?

Given its money-like characteristics, one top-down approach is to compare it to other forms of money. Depending upon the definition used, approximately USD 60tr of total global net wealth – estimated by Credit Suisse[12] to be approximately USD 150tr – is held in the form of money, implying a cash-to-net wealth ratio of 40%. Even when all 21 million Bitcoins are available (something that will not occur until 2025 – see footnote 6 above), at current prices the market capitalization of Bitcoin would be equivalent to less than 0.1% of all outstanding money and a meagre 0.035% of global net wealth.

Given the aforementioned benefits of virtual currencies we consider these percentages to be extremely low in an age where technology plays an increasingly dominant role in everyday lives – as we noted in our previous Market Insight. To us a more realistic level for this ratio is around 10% of all money being held in virtual form (implying roughly 2.5% of global wealth). Bear in mind there is an upper limit on this ratio because governments will be loath to give up the substantial benefits from seigniorage and hence will insist only government issued money can have full legal tender status.

Obviously, Bitcoin is only one of potentially many virtual currencies, and as we noted previously, there are strong motives for other virtual currencies to be created [13]. However, there is natural limit to the number of Bitcoin competitors (ie. a decentralized supranational structure) because money has strong networks effects.

Achieving a 10% virtual/total money ratio would require 100 virtual currencies all of the same size as Bitcoin is presently. Anything less than this number implies that Bitcoin’s price is, in a fundamental sense, too low. Presently there are over 700 virtual currencies. However, only three – Bitcoin, Ethereum and Ripple – account for over 80% of the total USD 60bn market capitalization for virtual currencies globally[14] – a clear indication of the strength of the network effect.

If in the end only ten virtual currencies dominate globally, and all have the same market capitalization (admittedly these are heroic assumptions but we are in unprecedented territory) the implied valuation for Bitcoin would be USD 29,000 per coin. Even taking into account the recent price rally[15], it would be massively undervalued and not, as the author of the blog post suggested, overvalued to the tune of 200%.

A valuation gap anywhere near this magnitude obviously supports the critique that Bitcoin’s use as a medium-of-exchange will be impeded[16] because it makes it a very attractive store-of-value. However, given the potential price upside it is really a strong enough of an objection to owning it? We don’t think so and neither do many other Bitcoin advocates!

Are our back-of-the-envelope calculations correct, or is the other author correct? No one knows and, as the saying goes, only time will tell. What is important though is that there is considerable uncertainty as to what Bitcoin’s fundamental value is and by definition a bubble can only occur when an asset price is significantly overvalued.

This is not the only reason for doubting Bitcoin is in a bubble, even though the recent exponential price gains would appear to support that conclusion. Just because an asset’s price rises sharply over a short period of time is not necessarily a reliable signal of a bubble. Consider the following exhibit, which shows the evolution of Bitcoin’s price over selected timeframes – obviously chosen to highlight our point – during the past five years.

Exhibit 2 – A Bitcoin Price Bubble – Now? …Now? …Now?

Source: www.quandl.com

During this relatively short time-frame there have been three occasions when Bitcoin’s price has risen exponentially and hence displayed price dynamics that at the time could be (were!) considered “bubble-like”. Even though significant corrections occurred after both of the first two price surges the simple fact is that from the first price peak of just over USD 200, it went on to hit USD 1,000 during the second price surge and, after the most recent price surge, stands at over USD 2,400. This is the problem attempting to identify bubbles based on price dynamics alone.

A major reason why Bitcoin is so susceptible to such high price volatility is there is no centralized issuing authority capable of smoothing out potential supply and demand imbalances. This means Bitcoin’s price has to continually adjust to bring the market back to equilibrium. With relatively small numbers of users, such price adjustments can therefore be quite dramatic implying substantial volatility. However, as the number of Bitcoin users increases, and the pool of buyers and sellers becomes more diverse, its price volatility will naturally tend to decline. Until this occurs, Bitcoin’s price will be particularly susceptible to the vagaries of its users’ collective mood. Indeed, we can illustrate this sensitivity by comparing crowd sentiment towards Bitcoin and its price – see exhibit below[17].

Exhibit 3: Crowd-sourced Sentiment vs. Price – Bitcoin

Source: www.quandl.com and www.amareos.com

Not every peak in crowd-sourced sentiment towards Bitcoin coincides with a price spike (there was marked divergence between the two series in late 2014 when crowd sentiment towards the virtual currency rebounded from record lows, positivity that was not reflected in Bitcoin’s price). However, it is true that both of the two earlier price spikes occurred after a sustained period of crowd positivity. What is particularly notable about the latest price surge is that just a few months ago crowd-sourced sentiment towards Bitcoin was negative and it has only risen modestly to levels that are hardly extreme relative to history and below that seen in the earlier two price spikes. This is not exactly the hallmark of irrational exuberance or hype that one typically associates with an asset price bubble on the cusp of bursting.

Hence, even if not for the faint hearted given its considerable price volatility and the potential for a meaningful price pullback, such crowd sentiment readings combined with uncertainty as to what Bitcoin’s fundamental valuation is leads us to reject the notion that Bitcoin is a bubble searching for a pin. Rather, the outlook for Bitcoin, especially over the medium-term, remains constructive, something many investors would question having witnessed the recent parabolic price gains.

_____________________________________________________________________

Amareos crowd-sourced sentiment indicators are based on MarketPsych Indices

FOOTNOTES:

[1] There is a serious, yet widespread, misunderstanding relating to Bitcoin’s lack of intrinsic worth. As we will argue, Bitcoin has many money features and not having any intrinsic worth is a virtuous characteristic for money not a defect. This is because if money – in whatever form it takes gold, paper, bytes – has intrinsic worth it is also be subject to nonmonetary demand and supply influences, which is far from a desirable property.

[2] Some companies do offer Bitcoin accounts that pay a rate of interest but this is really just repackaged credit risk associated with the margin traders who are borrowing the deposited Bitcoins – it is not an inherent feature of Bitcoin’s design.

[3] This is what is meant by a currency having legal tender status.

[4] Bitcoins are divisible up to eight decimal places.

[5] Visa typically processes 2,000 transactions per second while Paypal averages around 140 transactions per second. Data from “Modeling Decisions for Artifical Intelligence: 13th International Conference, MDAI 2016 – Proceedings”.

[6] The growth rate in the supply of Bitcoins will gradually decline until all 21 million coins have been mined. The supply of Ether – the currency of Etherum – is not capped.

[7] See: http://positivemoney.org/2014/04/bitcoins-fatal-design-flaws/

[8] More on this latter.

[9] This is a consequence of the widespread adoption of positive inflation targets – typically set around 2% per annum.

[10] See: https://ftalphaville.ft.com/2017/05/22/2189145/how-to-value-bitcoin-with-a-traditional-valuation-measure/

[11] Think: The Silk Road online black market.

[12] See: https://www.credit-suisse.com/uk/en/about-us/research/research-institute/global-wealth-report.html

[13] Aggregating the market value of all crypto-currencies comes to USD 60bn, of which Bitcoin accounts for roughly 50%. (It was a high as 75% at the start of the year).

[14] See: http://www.economist.com/news/finance-and-economics/21722235-bitcoin-far-only-game-town-surge-value-crypto-currencies

[15] The next time some argues that Bitcoin, or any of the other virtual currencies, are seriously overvalued and in a bubble, it might be worth going through the above valuation exercise with them and see if their attitude changes. NB: If Bitcoin were to be the only virtual currency, and it were to fully replace fiat money, each coin would be worth an eye-watering USD 2 million.

[16] At least until such time as the perceived valuation gap is closed.

[17] Big thank you to Richard Peterson and Alexey Karakulov for the data, which is still in development. In the exhibits we show the combined sentiment readings for mainstream and social media. We did check each individually but, and much to our surprise, we did not observe a marked difference in crowd attitudes between the two media types. (We had expected social media to be much more positive reflecting the almost religious-like faith of tech-savvy Bitcoin believers).

Article Topics

Get In Touch

Subscribe

We have updated our Privacy Statement. Before you continue, please read our new Privacy Statement and familiarize yourself with the terms.x