Friday 28 July 2017

Why I think digital currencies will go to zero

In 2011, my girlfriend at the time, who knew next to nothing about finance and investments, asked me out of the blue if she should buy some gold. She produced a bouchure she had procured from somewhere offering for sale small ounce-sized gold bricks. I was already bearishly inclined towards the metal at the time, but that comment cemented my bearishness. It called immediately to mind the depression-era saying that "when the shoeshine boys start telling you what stocks to buy, it's time to sell" (paraphrased). Sure enough, gold, which was at about US$1,800/oz at the time, was within a hare's breath of peaking. Gold now trades at closer to US$1,200/oz - a 33% loss over a 6yr period where the S&P500 has continued to surge to record highs.

An interesting dejavu moment occurred recently, when several old friends of whom I am in irregular contact, who are not involved in finance, all messaged me at about the same time earnestly asking for my opinion on Ethereum, the latest digital-currency fad du jour. I had the same instinctive reaction as I had to my girlfriend's inquisitions about gold - run! So far that instinctive judgment has proven correct. Indeed, Ethereum's price peaked within a few days of receiving those messages, and has now already declined by nearly 50%, from a shade under US$400, to nearly US$200. The market capitalization of all Ethereum coins outstanding still remains at about US$20bn, however - a not inconsiderable sum.

For the record, I think Ethereum and all other digital currencies likely eventually to go zero, or something very close to zero. In what way shape or form, however, I don't know. The madness of crowds cannot be predicted in advance via rational analysis, and pricing outcomes are path-dependent in nature. It may take a detour much-much higher. But I believe something approximating zero is the ultimate destination.

Furthermore, I have a hunch (albeit only a hunch) that we may already have seen the highs, for the same reason gold peaked in 2011 at the time my girlfriend expressed buying interest. Speculative booms require fresh oxygen all the time, and that oxygen is supplied by new buyers. When the average non-financial man-on-the-street type is already bulled-up and buying, you know that the point where the supply of new buyers is nearing exhaustion is already close at hand. It is the same reason ponzi schemes all eventually burst: the supply of new buyers is eventually exhausted. And when prices start to fall, the psychology rapidly changes, with greed quickly morphing into fear.


Fundamental fallacies

Aside from the typical psychological fallacies associated with all speculative bubbles, every bubble has at its core some fundamental misperceptions. In the case of digital currencies, there are several in my opinion, five of which I discuss below. These fallacies have provided sufficient rationalization for many participants to engage in what would otherwise transparently amount to little more than a speculative orgy

The first fallacy, at least as far as Bitcoin is concerned, is the idea that because the number of coins is finite (capped at 21m), that this somehow imbues the coins with enormous value. The finite number of coins, it is argued, represents a vast improvement on fiat currencies, where the supply is subject to the whims of central bankers, and is typically always on the increase. Ergo, Bitcoin should represent a vastly superior store of value over the long term than fiat currencies.

The problem is this: while the supply of bitcoins might be finite, the aggregate supply of digital currency as a whole, including Ethereum, Litecoin, and all the other variants, is potentially infinite. We have already seen the emergence not just of Ethereum, but a seemingly limitless number of new Bitcoin immitation currencies alongside a proliferation of so-called 'initial coin offerings'. It is therefore a clear fallacy that one is buying into a finite supply of the world's next global currency to be.

And why should we not have expected supply to mushroom in this way? Ethereum was created by a 21yo college dropout in his basemen at zero cost. If you can code a currency in your basement that subsequently ends up trading at an aggregate value of $40bn, then why on earth would you not do so? But this sort of alchemy can't last. One of the best measures of something's value is its replacement cost, and if a 21 year old can create a digital currency in his basement, it can't be all that hard. It is therefore reasonable to expect thousands and thousands of new coin offerings to emerge until the supply overwhelms demand and the prices of all of them crash.

This is the same dynamic that eventually contributed to the dot.com crash in 2000-02. The booming share prices of worthless dot-com start-ups incentivised investment bankers to bring as many new IPOs to market as possible as fast as possible. As they did, the supply of worthless scrip continued to grow and grow until it eventually overwhelmed what previously appeared to have been an insatiable degree of speculative demand. At that point prices started to fall, and then the psychology rapidly changed, with everyone simultaneously rushing for the exits. Prices crashed and most market participants were wiped out.

In the case of Ethereum, not even the supply of Ethereum coins is fixed. Instead, new coins are constantly issued to network participants that host the currency and process transactions. The platform is therefore ponzi in nature and relies on a continuing increase in coin supply (hardly something that speaks to the long term store-of-value utility of the currency). In Bitcoin's case, it is also unclear who is going to supply the computing power to host and process transactions once the entire 21m of bitcoins are 'mined out' - a potentially fatal ponzi element of the currency.

Central banks can indeed be faulted for willingly increasing the money supply to foster their inflation targeting goals (typically around 2% pa), not to mention having engaged in an irresponsible (in my opinion) degree of quantitive easing in recent years. However, the potential supply of digital currencies is infinite, and is growing faster than even the Venezuelan money supply. And anything whose supply is increasing exponentially will eventually see its price trend towards zero - particularly something where what you are buying is literally nothing more than ones and zeros on someone's server.

The second and related fallacy is the idea that if blockchain technology is indeed revolutionary and is eventually widely adopted as a means of processing payments, then existing digital currencies will end up having a lot of value. There is no reason to think that they inevitably will in my opinion, because as discussed, new currencies and blockchain platforms can be conjured at will. There is no reason existing financial institutions could not get together and launch and host their own blockchain platform, for instance. Blockchain might prove revolutionary for payments, and digital currencies might still end up being worth zero or something close to zero.

Thirdly, there is a fallacy that a digital currency can act as a reliable 'store of value'. It cannot, because all you are buying are ones and zeros on someone's computer, that have zero intrinsic value, and only have exchange value so long as somebody else is willing to buy them off you. A tulip in the midst of the 16th century tulip bubble had exchange value, but almost no intrinsic value, and buying a $10k tulip for its 'store of value' merits would have proved a very bad idea.

Because the only value digital coins have is their potentially ephemeral exchange value, they are therefore at constant risk of being rendered worthless if people are no longer desirous of holding them. This could happen for any reason at any time. In addition, prices are so volatile that they cannot reasonably function as even a reliable means of exchange for anything other than immediate transactions of trivial size. If you sold you house, would you agree to take settlement in six weeks time in Bitcoin or Ethereum? If you had done so with the latter several weeks ago, a house sold for say $1m would now yield only $500k. The degree of uncertainty as to the coins' future value over timeframes as short as even one to two months is too large for these coins to function as practical means of exchange.

It is true, of course, that fiat currencies also have no intrinsic value. However, that does not mean they are equivalent to digital currencies in any meaningful way, because (1) fiat currency is generally not held in large quantities as a store of value - it is held primarily for its transactional value; real wealth is usually held in stocks, bonds, real estate, etc; (2) cash deposits have the capacity to pay interest, so if inflation picks up, interest income can offset the inflation to some extent, preserving its real value; and most importantly (3) existing fiat currencies are in entrenched use (due to both  legislative monopolies, and network effects), and their medium term transactional value and widespread acceptance is therefore highly predictable. This cannot be said of digital currencies.

Fourthly, there is a fallacy in the idea that new currencies are needed to provide new payment systems/solutionsCurrencies and payment systems are not the same thing. The world does not need new currencies - what it needs is improved payments systems that are lower cost and more frictionless. They are already coming down the pike. Digital wallets and new and improved electronic payment systems are proliferating everywhere, from Apple Pay and Samsung Pay, to Alibaba's Ant Financial, which is growing like a weed. Even AirAsia is looking into launching a 10-currency digital wallet. This is the way of the future - cheap, frictionless, and widely adopted digital payment systems that utilize existing currencies, not new faddish digital currencies.

Getting a new payment system widely adopted is already extremely difficult, and getting a new currency adopted is even harder. It almost never happens. There is a good reason for that: our existing currencies and payments systems already do a reasonably good job. Physical cash works well for small payments - it is relatively frictionless and completely costless for users. Debit cards and credit cards also function pretty well, although the latter is more expensive than it needs to be and will probably remain longer term (interchange fees seem destined to be competed down in my view). Inflation rates are also low throughout most of the world at present, so existing fiat currencies are function very well as a store of value (and USD and other 'hard' currencies can be used as a substitute for official local currency in mismanaged developing countries with hyperinflation). Bitcoin is a solution looking for a problem that doesn't really exist, or certainly which can't be solved with the use of existing payment systems and currencies.

Lastly, the final fallacy is the idea that governments will allow a new global distributed-ledger currency to exist and enter into widespread usage without stringent regulatory oversight. This is extremely unlikely. Banks and the financial system are heavily regulated. Digital currencies are not yet similarly regulated because their use for anything other than speculation remains trivial, but if usage of digital currencies rises, they will likely be similarly regulated. Unfortunately, this would neuter a key part of digital currencies' value proposition - their distributed, decentralised nature.

The above fallacies have all been used to justify buying and holding digital currencies as an investment, or to rationalize what is really little more than thinly-veiled speculation. Unfortunately, none of these rationalizations bears scrutiny in my opinion, and I believe digital currencies to be nothing more than a faddish bubble.


That doesn't mean the currencies go to zero any time soon, or that speculation is irrational

The above, however, does not mean that the coins will inevitably go down from here, or indeed, will not rise significantly further before they go to zero. The market capitalization of Apple is about US$800bn. Bitcoin is 'only' about US$50bn (Apple of course produces US$50bn of free cash flow a year, whereas bitcoin is just worthless ones and zeros). There are trillions of dollars of bank deposits sitting idle that could flow into a speculative asset like Bitcoin, and history has shown that speculative extremes can reach levels that surprise most rational pessimists.

It is also not the case that speculation is necessarily irrational, contrary to what many believe. George Soros was the ultimate rational speculator, and said that the first thing he did when he saw a bubble developing was rush to buy. It is not irrational to speculate because you can only lose 100c in a dollar, but you potentially stand to make 5x, 10x, or 100x your money - at least temporarily - if you get in early enough and the bubble inflates significantly enough. There only needs to be a reasonable probability that a huge bubble develops for it to be a rational speculation. To some extent, this is the dynamic that drives bubbles - at least initially.

The smart speculator, however, knows what is going on, gets in relatively early, and knows that the boom will eventually turn to bust, and is therefore prepared for it (Soros liked to switch from long to short when he felt the market was rolling over, profiting from both the boom and the subsequent bust). Few people who speculate do this successfully; most get caught up in the hype, buy in late, end up averaging down as prices decline rather than selling/shorting, and end up selling at or near the bottom and losing almost everything.

I would counsel caution. But don't blame me if the coins double or triple on their way to zero.


LT3000

(Have never owned any digital coins, and never will for investment/speculative purposes)