Saturday 2 December 2017

Bursting the Bitcoin Bulls' Bubble

I really hate to do this. There has already been an inordinate wastage of humanity's collective time spilling ink on Bitcoin in recent months - something I contributed to with my first relatively-long post on the topic here (200% ago). But with the digital 'currency' having recently hit US$10,000/coin, I can't resist a follow-up. So permit me a brief indulgence.

No doubt Bitcoin's many owners are feeling quite chuffed with themselves right now. Their prescience, after all, has been vindicated by continuing strong price gains (in reality, all this proves is that there have been more buyers than sellers of late - not that the thinking behind the original purchase was accurate). Some may even look somewhat pitifully on old-schoolers such as myself, that have missed the boom and still just don't seem to get it.

I hate to burst the bulls' bubble with facts and reason, and be a proverbial buzz kill, but I'm going to attempt to do so. Let's deconstruct this US$10k/coin price for a minute. Based on the 21m coins that will eventually be outstanding, Bitcoin's recent 5-figure price gives the total amount of outstanding Bitcoin a combined market value (or 'market capitalisation') of US$10k x 21m = US$210bn.

Let's suspend disbelief for a moment, and assume that Bitcoin really does become a future means of exchange in widespread (nay, global) usage. What sort of valuation would this justify for Bitcoin? If widespread transactional usage were to eventually occur and determine Bitcoin's price, its equilibrium market value would be a function of two things: (1) the total number of users with a digital Bitcoin wallet; and (2) the average value of coins held by the average user for transactional purposes (in much the same way you maintain physical cash in your wallet, or weekly float in a chequing account). You would multiply the two numbers together to get the aggregate value-in-use of total Bitcoins outstanding.

To justify Bitcoin's current price (let alone all the other cryptocurrencies currently in existence - including Bitcoin 'spin-offs' such as Bitcoin Cash - as well as the countless future currencies that are sure to be conjured from the ether via ICOs), if every man, woman, and child on the face of the planet were to hold a Bitcoin wallet (all 7.6bn of us), they would need to hold an average of US$30 worth of Bitcoins each. If instead, merely every man woman and child in the developed world (1.2bn) had a wallet, they would need to hold an average of US$175 per wallet. On a per-household basis, the numbers would rise to an average of about US$100 and US$500, respectively.

These are not negligible amounts of money per wallet even based on an implausibly optimistic, best-case-scenario estimate of future user numbers. Furthermore, that is simply the amount required to justify Bitcoin's current price. Most people are buying Bitcoin because they want/expect to make at lot of money, and fast. Let's say you think the price is going to go to US$100,000. Well, that price would require every person on the planet to have an average of US$300 worth of Bitcoins in their wallet (US$1,000 per household), or 1.2bn people to have US$1,750 worth (US$5,000 per household), or some alternative combination thereof.

These figures are all orders of magnitude larger than the number of PayPal accounts (and associated account balances) that exist some two decades after the company's founding, and PayPal is by far the world's most successful new payment network in the past 20yrs. Does that mean the Bitcoin penetration numbers highlighted above are impossible? No it doesn't. But such outcomes are exceedingly unlikely in my view, given the considerable practical barriers to widespread adoption that exist, coupled with the highly effective job our existing currencies already do in facilitating transactions (see my original post).

The best that can be said is that one is merely buying into the (vanishingly small in my view) possibility/probability that Bitcoin eventually achieves the level of success discussed above - i.e. a highly speculative long term call option on that possibility - one that will most likely prove worthless, but which has a small probability of a large payoff.*

How much is that option worth? Let's assume (exceedingly generously) that the probability of success and widespread adoption is 10%. A 10% probability of success requires all of the above usage metrics to be multiplied by a factor of 10x to derive a probability-weighted fair valuation today, and that valuation also ignores the time value of money (it will take a long time for such widespread usage to occur even in the best of cases - likely 10yrs at a minimum). That hardly seems like a uniquely attractive risk/reward proposition.


It potentially gets much worse, however

However, the biggest problem of all, in light of the massive usage numbers now required to justify Bitcoin's monstrous valuation, could well be the very nature of the Bitcoin blockchain technology itself. Unless existing problems are solved in the future, Bitcoin already appears doomed and worthless.

Now, I'm not going to pretend to be a computer science expert that can speak authoritatively on this topic, so I may be wrong and may have missed something. However, numerous sources note that Bitcoin's blockchain foundation suffers from a significant scalability problem that severely limits the number and speed of transactions that can be processed. Wikipedia, for instance, reports that the limitation of Bitcoin blockchains to 1MB caps the number of transactions at about three transactions per second. This is not materially out of line with recent Bitcoin transaction rates shown in the second link below.

https://en.wikipedia.org/wiki/Bitcoin_scalability_problem

https://blockchain.info/charts/transactions-per-second

If this in fact proves to be an intractable issue, then Bitcoin has a serious problem. Let's return to our hypothetical example where 1.2bn users held an average of US$175 per wallet - a level of usage/adoption already required to justify Bitcoin's current price. Let's say these users wanted to use Bitcoins to execute just one transaction per person per day on average (considerably less than the typical number of transactions people will undertake on a daily basis) - perhaps buying lunch or a Starbucks coffee. Just one transaction per person would require 1.2bn transactions to occur per day.

The problem is that there are only 86,400 seconds in a day, which at a processing rate of three transactions per second, would result in transaction capacity of only 259,200 transactions per day. That is somewhat short of the requisite 1.2bn. At base, what this means is that the Bitcoin's existing blockchain technology is simply not capable of processing remotely enough transactions in real time to support the extent of the user base already required to justify Bitcoin's current exorbitant price tag.

Furthermore, as I discussed in my original post, the copious amount of computing power currently being used to process the existing meager number of transactions (which have already created severe difficulties and delays in trading Bitcoin), is being supplied by 'miners' that are only supplying such expensive and energy-sapping computing power because they are being rewarded with newly-minted Bitcoins. To date, this has not been a problem, because Bitcoin's spiraling price has continued to justify an increased pace of mining (and successfully offset the impact of the steady reduction in the pace of coin issuance - a necessary tapering to keep the total issuance capped at 21m coins).

However, if one cares to stop and think about it for a moment, it doesn't require an Einsteinian intellect to see that the entire structure of the Bitcoin transaction processing infrastructure is nothing more than a ponzi scheme writ large. This is because the continuity of the platform's transaction processing operations relies on a growing stream of computing power to process a growing number of transactions over time, but this power can and will only be supplied if the reward for mining exceeds the cost of doing so. And with a steadily declining pace of new coin issuance, this can only be achieved if Bitcoin's price continues to spiral forever higher.

If, by contrast, the price of Bitcoin were to suddenly collapse to a point that no longer justified miners expending vast amounts of computing resources and costly electricity to 'mine' new Bitcoins, it is possible that transactions in the currency could suddenly grind to a complete halt without any warning, with a processing backlog suddenly emerging that spans many years, as the system is flooded with sell orders and miners disappear. Many Bitcoin owners could suddenly find themselves completely unable to withdraw any their funds (rather than fall, the coins may simply cease to trade, resulting in a complete loss). Confidence would likely evaporate overnight; buyers would disappear; and the price of other cryptocurrencies that were still trading would likely instantaneously collapse 80-90% or more.

Of course, it is possible the software might be able to be improved in a manner which overcomes this structural limitation. This is what is behind the various Bitcoin 'splits' being discussed, and the growing disharmony amongst Bitcoin developers. However, the best way to do this might be to re-write the code entirely. Let's say I was a gifted young programmer in a garage somewhere, and that I came up with a new code that was a vast improvement over Bitcoin, and was capable of processing a significantly increased number of transactions with improved efficiency. I could use this new and improved code to launch a new digital currency called 'LyallCoin'. Suddenly, Bitcoin could find itself obsolete and worthless, having been superseded by a better code (digital currency owners are still yet to level with the reality that computer algorithms can be copied ad infinitum).

Alternatively, it is possible that the blockchain processing limitation is structural and incapable of being fixed, and that the only way to process a large number of transactions rapidly and over an indefinite timeframe is to engineer the currency to continually issue more and more units. This would inevitably depreciate each and every viable digital currency towards zero over time - and probably at a rate much faster than most fiat currencies lose there value. Time will tell, but if I was holding Bitcoin, I sure as hell would want to be confident that the currency's prospective viability had at least been established (to use Elon Musk's parlance, I would want to be confident that 'success is one of the potential outcomes').


So when does it all turn to pumpkins and mice?

The reality is that none of the above is likely to resonate with any of the Bitcoin bulls or in any way alter behaviour, because the bulls are are not buying Bitcoin because they have any real understand of these issues; understand a thing about how blockchain technology actually works; or even care about the coin's long term viability. They are buying Bitcoin for the same fundamental reason people always participate in speculative bubbles: a desire to get rich quick without work.

Human beings' brains are pattern-recognising machines, and when inexperienced investors see prices going up and up, they think they see a pattern. Writ large, this belief becomes self-fulfilling, and over time through a process of continual positive reinforcement, as prices continue to rise, buyers come to form a strongly-held emotional conviction that prices are sure to rise, even when that emotional conviction is completely divorced from economic reality. Furthermore, this emotional conviction can become so strongly held as to render belief completely immune from rational argument.

Humans are also wired through evolution to be highly-efficient social learners capable of picking up on subtle social cues from their surroundings (often subconsciously), and are therefore readily influenced by the behaviour and collective belief of their peers. Combined with intensely-held emotional convictions about the inevitability of price rises, this sense of social proof and solidarity also supports the bubble, until it eventually collapses under its own weight. These are the real factors driving Bitcoin. It has nothing to do with digital currencies. Bitcoin is merely the latest iteration of a pattern of human psychological fallibility that has manifested itself repeatedly throughout history.

This is far from the first time a speculative mania has swept up the world, and it won't be the last, because human beings' greed, innate laziness, and credulity are hardwired and incurable. The current Bitcoin bubble has much in common with the 17th Century Tulip Mania, where at its height singular Tulip bulbs sold for more than entire houses. Viewed rationally from a safe distance, participants appear almost unbelievably foolish, but the psychology that afflicted these punters is little different to what we are seeing at present in the world of Bitcoin. The world has learnt a lot since the 17th century - the science of electricity; telephony; and microprocessors, to name a few - but it has learnt relatively little about the dynamics of speculative manias (progress in science is cumulative, but progress in finance is cyclical), and a new generation of investors looks set to imminently learn some very old lessons.

When? I still have no idea. However, as I noted in my original digital currency post, a speculative bubble relies on a continuing supply of fresh oxygen to sustain itself. That increased oxygen has been in visibly growing supply since my original post. For instance, it can be readily observed that the frequency with which Bitcoin is now discussed in the popular press has continued to increase, as has the level of popular participation and interest in the topic. Two out of the top 10 stories in the Wall Street Journal today were on Bitcoin. That was not the case three months ago.

There is still room for the currency to go considerably higher, but it is clear we are getting closer to the point where the supply of fresh oxygen (read, gullible victims being drawn into the speculative bubble late in the cycle that are destined to lose all their money) will start to be exhausted. At this point, price declines are apt to be swift and severe. Time will tell how and when this plays out, but I for one am standing well clear.


LT3000

*Incidentally, one of the more ridiculous justifications I have heard for owning Bitcoin is that it is a 'store of value'. Bitcoin is an experiment without precedent, and like all novel experiments, there is a substantial risk of complete failure. To call any such speculation with a high probability of a complete wipe out a 'store of value' is absurd. What people are really saying is that 'in the past Bitcoin has acted as a store of value'. But that can be said of any investment that went up in price in the past - it was also true of Madoff's investment fund when looking backwards as well. 

For something to be a store of value, there has to be a strong argument for why the investment can be reasonably expected - come hell or high water - to preserve its value in the future. There is absolutely no basis for that belief when it comes to Bitcoin.