Thursday 21 December 2017

Bitcoin addendum: Running out of oxygen; the 'money of the internet' fallacy; and bitcoin futures & systemic risk

I really can't resist a quick addendum to my recent bitcoin post. I thought I had said all that needed to be said. But as Bitcoin's price has escalated to new highs of nearly US$20,000, there has been a commensurate rise in the degree of folly and fuzzy thinking, and I can't seem to keep my thoughts on the matter to myself. I have three additional points/observations to make:


#1 The supply of fresh oxygen looks to be nearing its limits

I commented earlier this month that 2 out of the top 10 Wall Street Journal stories were on Bitcoin (or cryptocurrencies generally) - up from merely an occasional article 6-12 months ago. Well, yesterday, 3 of the top 9 stories were on cryptocurrencies (and predominantly Bitcoin), and everyone - and I do mean everyone - is now talking about Bitcoin. If we are not already at the top, we must be damn near close (temporally; a final euphoric spike is always a possibility). I will be very suprised if the cryptomania does not top out in early 2018, if not sooner.

The point to consider here is that any asset is going to have a natural level of daily selling reflecting some small fraction of its aggregate market value. Some minority of the holders of an asset will decide to sell on any given day for a variety of reasons. These may include the need for money for personal reasons, or because they decide to take some profits or rebalance their portfolio, etc. Liquidity in the housing market works in a similar way - there will always be some natural level of selling from the liquidation of estates; people moving or moving overseas, etc.

The market capitalisation of all Bitcoin was recently approaching US$400bn, and let's assume for the sake of argument that the average natural level of selling is 25% of the total issuance each year (i.e. the total Bitcoin stock turns over about once every four years - an assumed holding period likely substantially in excess of the actual average). That means that US$100bn of new buyers have to be found every year, or US$274m a day. Let's round it to US$300m a day.

That US$300m means that if the average new buyer purchases say US$1,000 of Bitcoin, that 300,000 new buyers suckers need to be found every single calendar day to absorb the flow of natural selling merely for the price to not go down. For the price to rise, an even greater level of buying need occur. And as the price rises, the aggregate dollar value of this steady flow of natural selling correspondingly increases. For instance, if Bitcoin's price was to double again, 600,000 new daily buyers of US$1,000 worth of Bitcoin would need to be found in order for the price to not fall, and so forth. 

This is what it means for a bubble to eventually collapse under its own weight. As the price rises, the number of new buyers needed continues to rapidly rise, whereas the number of potential untapped buyers continues to shrink (because you increasingly reach a point where most people that are willing and able to buy the asset have already purchased it). When an inflection point is reached, where the follow of natural selling begins to exceed the pace of influx of new buyers, the price will start to fall, and when it does, market psychology also changes, causing buyers to pause and the pace of selling to increase, as greed quickly turns into fear.

Whether we have reached or are about to reach this inflection point with Bitcoin remains to be seen, but I wouldn't be at all surprised if we had. Everyone is talking about it and it is all over the media, and prices have recently gone parabolic, which have all historically been reliable contrary indicators that suggest that the pace of new buyer influx is likely at or close to its high. Bitcoin is about as close to a textbook bubble you will ever find, and such bubbles almost never end well for the longs.

#2 Money for the internet fallacy

A recent media article on the 'Bitcoin Billionaire' Winklevoss twins quoted them as saying that Bitcoin was the first money "built for the internet". Commenting on the slow, cumbersome, and costly process involved with traditional international money transfers intermediated by conventional banks, they noted:

"The joke is if you want to get money from New York to London on a Friday night, go to JFK, jump on a plane with a bag of cash and you'll get there quicker than when you actually wire. [Bitcoin] is the first money that's built for the internet".

With this comment, the Winklevoss twins are doing little more than demonstrating their profound naivety and ignorance about how the financial system works. The considerable delays often associated with international money transfers have little to do with technology or the limitations of our existing currencies, but instead reflect the significant regulatory and compliance burdens banks operate under, including the need to comply with mandated central bank processes for dealing in foreign exchange markets. And governments never move quickly. There are also checks and balances in place to prevent fraud and errors, which are important when large sums of money are involved.

Banks are subject to all kinds of onerous regulations, including the necessity to undertake extensive KYC (know your customer) due diligence, and monitor transactions to help detect, report on, and prevent money laundering, terrorist financing, and fraud, amongst other things. In addition, many central banks tightly monitor and regulate foreign exchange market activity, due to the systemic importance of exchange rate volatility, and foreign currency transactions in many countries require considerable paperwork to comply with stipulated central bank procedures and hard and soft currency controls. And these actions need to be co-ordinated across boarders with banks operating under different regulatory regimes. It is these regulatory factors that are the cause of delays, and they have nothing to do with existing currencies being less 'digitally enabled' than their crypto counterparts.

For a more realistic flavour, consider the following exerpt from a recent article discussing why Malaysia's central bank recently banned all deposits and withdrawals from Bitcoin wallets until dealers furnished the CB with increased data and transactional transparency. The article noted:

"Among the data needed by the central bank [before bitcoin wallets are allowed to operate again] are the number of transactions, including the conversion from digital currency to fiat money and vice versa as well as conversion from digital currency to another; the value of the transactions; net buy and sell position; purpose of transactions, payment method and the number of customer accounts."

The reason transactions are slow is because banks are obligated to meet all these obligations. If they were not subject to these restrictions, transaction speeds would be much faster and more frictionless. Indeed, that is already the case for many regular transactions. For instance, regular ATM cash withdrawals; the processing of small debit and credit card transactions; e-money transactions; and online or mobile domestic money transfers, are all extremely fast, reliable, low cost, and secure - indeed much faster, much cheaper, much more reliable, and much more secure and fraud-free, than cryptocurrencies, which are an inferior solution (this is before discussing the considerable problem of the lack of security and hackability of Bitcoin wallets - US$70m worth of coins was stolen just over a week ago; if you thought you held the coins, tough luck).

The bulls may attempt to argue that the whole attraction of Bitcoin is that it is unregulated and distributed in nature and therefore bypasses the need for all this regulatory and compliance burden. But again, it is naïve in the extreme to assume that regulators will willingly allow their powers of monetary and financial system management to be usurped, or be powerless to intervene to prevent that outcome. In my opinion, there is simply no way regulators and central banks are going to allow unmonitored and unregulated transactions to occur in scale via cryptocurrencies, and as we saw in Malaysia this week, if they wish to intervene to prevent that from occurring, it is a simple exercise to ban all banks from accepting deposits or withdrawals from bitcoin wallets forthwith, rendering the so-called currencies unusable in the said country overnight.

More of this regulatory clampdown can be expected to happen over time if cryptocurrencies continue to grow. In the past they were mere curiosities, but that will change if they grow and become more systemically important. Furthermore, from the perspective of modern monetary management, they are highly flawed constructs - their putative limited supply and rising value are inherently deflationary in nature, and hence anathema to modern central banking (be careful what you wish for - a highly deflationary currency will crash the economy - a currency that slowly loses its value over time is much more conductive to economic activity; the gold standard failed miserably and was scrapped partly for this reason). Even if Bitcoin were a good idea in theory, in practice it will never be able to function as the frictionless global currency the bulls hope it can and will be able to.


#3 Bitcoin futures make their debut, potentially ushering in growing systemic risks

To much fanfare, both the CME and the CBOE have recently launched Bitcoin futures products, and several investment banks/brokers have noted that they will allow their clients to trade them. The bulls view this as a sign that Bitcoin has now started to go mainstream, and the apparent stamp of approval shown by major traditional financial industry players is seen as proof of Bitcoin's growing legitimacy. 

I know enough about the financial industry - having observed it over many years, and having worked for an investment bank in a past life for over a decade - to know that their willingness to trade Bitcoin-related financial products says nothing more than that they think there are fees to be made trading these products in the short to medium term. At best, they will be agnostic on the coin's investment merits; at worst they will fully expect the digital currencies to eventually go to zero, but hope to make money trading the products in the interim.

The reason futures products have popped up is simply that Bitcoin has now grown to a size where its total market capitalisation is approaching US$400bn, such that the trading of financial products related to this new 'asset class' is now a major revenue generation opportunity. However, that absolutely should not be viewed as providing any sort of assurance of Bitcoin's legitimacy. The banks will make and keep their trading income even if Bitcoin eventually implodes. Indeed, historically, the financial industry is far from having covered itself in glory through the exercise of restraint in choosing to forego short term fee opportunities due to the long term financial risks they entail.

Furthermore, while many see the arrival of Bitcoin futures as, as worst, a curiosity, I actually see something a lot more sinister and worrisome. Bitcoin is not like most financial products against which derivative products are traded, in that most underlying instruments like commodities or stock indexes, have intrinsic value and an inherent limit on the realistic degree of volatility the underlying instrument is likely to experience. However, as I discussed in my prior post, Bitcoin has the realistic risk of seeing trading completely seize up and the 'currency' essentially go to zero overnight, and this is a risk that seems to be poorly understood by most market participants, who instead appear eager to liken it to any other financially-tradable product or 'asset class'.

This was not such a big deal when Bitcoin was a relatively small cash-settled curiosity. If you were long Bitcoin and it suddenly collapsed to zero, it would hurt, but its failure would not have systemic consequences. However, if Bitcoin futures products grow to any appreciable size, this will change, because suddenly large amounts of Bitcoin futures contracts trading on margin could experience a dramatic change in value, and the amount of underlying margin could prove woefully inadequate. We saw the consequences of this in the Global Financial Crisis, when large derivative bets against the decline in the value of tranches of mortgage bonds imploded, bankrupting several institutions such as AIG.

Counterparties at least have recognised the elevated risk of outsized volatility by offering a maximum of 50% margin (vs. allowing as much as 96% margin on daily positions in conventional currencies and commodities). However, if the amount of outstanding notional Bitcoin futures were to grow to a material size of say 1x the underlying instrument, or US$400bn, on an average margin of 50%, up to US$200bn of financial system capital could be at risk - enough to have systemic consequences - and worse still, competitive pressures may also erode the size of this safety margin as well. If the coins were to suddenly and unexpectedly go to zero, this could hand important leveraged financial market counterparties serious, life-threatening losses on defaulted futures contracts that its customers were unable to settle. That could be potentially destabilising to the financial industry (Buffett, let it be recalled, famously described derivatives as financial weapons of mass destruction).

My broker (and the primary custodian on an upcoming fund I am launching), Interactive Brokers, for instance, is offering Bitcoin futures. A few years ago, IB lost more than US$100m after the Swiss Franc was unexpectedly allowed to sharply appreciate by the Swiss central bank. It rose about 20% in a day, which resulted in many futures contracts having inadequate cash collateral to settle, and defaulting. IB has some US$5bn in capital so was comfortably able to absorb the hit, but it does highlight the risks posed to counterparties offering access to leveraged derivatives products when an instrument experiences an unanticipated and outsized degree of volatility.  

For now, the amount of outstanding Bitcoin futures issuance is small, and so we need not worry - particularly if the bubble is already cresting highs and about to imminently implode. However, if the bubble keeps inflating and the degree of outstanding notional Bitcoin futures increases, and competition between providers drives down the level of required margin, then we will need to be very careful, because unfortunately, the track record of the financial industry of foregoing a short term fee generating opportunity in order to guard against longer term systemic risks is not encouraging, and the industry has a long track record of successfully inventing new ways to bankrupt itself.


LT3000