## Tuesday, November 20, 2018

### Aspects of the bursting cryptobubble

Eight days ago, when I wrote the most recent Bitcoin post, the Bitcoin price was still stuck close to $6,350 – a plateau that had been almost stable for more than a month – and I have reminded the readers that such a stability was unsustainable and the price was guaranteed to drop further. We're around$4,600 now – a 28% drop in those 8 days – and the price has already looked below $4,500 which is 77% below the all time peak of$19,800. Almost all (unpegged) cryptocurrencies were falling in the recent week – the total capitalization of them is around $150 billion. For some reason, Ripple has partly avoided the downturn while Ethereum suffered more than others. The newest Bitcoin Cash fork was quoted as a cause of the anxiety – well, I think it is way too oversimplified to find such a simple scapegoat. Two different reforms of the Bitcoin Cash (which is the most influential fork of the Bitcoin itself) rules were proposed – for the (former) #4 currency to become Bitcoin Cash ABC (Roger Ver and Jihan Wu, one of them is modestly called Bitcoin Jesus) and/or Bitcoin Cash SV (which stands for Satoshi Vision because the backer, Craig Wright, has at least called himself Satoshi Nakamoto once). The "hash war" – which wasn't really just a hash war and the rules seem extremely uncontrolled to me – was concluded with the victory of ABC. Crypto folks were used to the (irrational) free dividends whenever there was a fork. This time was different. The$400 Bitcoin Cash turned into a $220 Bitcoin Cash ABC plus$50 Bitcoin Cash SV – both exist at this moment. So in total, they lost about 30% like almost everyone else. Stellar overtook Bitcoin Cash as the #4 currency after the Bitcoin, Ripple, and Ethereum.

Now, if you look at recent BTC blocks at btc dot com, you will see that the mining hasn't slowed down substantially at all. Ten blocks are produced roughly in 100 minutes (1:40:00), just like the plan indicates. There is a widespread belief that most of the miners must already be making a loss. But they probably continue because if they stopped mining, the belief system underlying their whole life would crumble.

The dollars from the newly mined Bitcoins almost certainly fail to pay both the electricity and the hardware investment. But even if you accept this kind of a loss as unavoidable, there's still the question whether you should keep on running the mining rigs. And this question is answered by looking whether the newly created Bitcoins pay at least for the electricity itself. I think that even this outcome might be too optimistic.

In the left lower corner of the "btc dot com" page, you see the most productive mining pools. "Btc dot com" itself (a Chinese rig) is at the top but you see something interesting on the second place. SlushPool, the world's oldest Bitcoin mining pool run by Czechs, is there. That indicates a self-evident growth in importance in recent weeks – for most of the year, it was around the 4th or 5th place.

Now, I would say that SlushPool is the "mainstream mining pool for the Western public". Who are the members? I think that they are mostly kids using the good old GPUs (graphics cards). The economy of this mining is subtle, of course. Kids mine $10 in Bitcoins through this mining pool while their parents pay$50 for electricity and $200 for hardware. The total expenses are$250 and the revenue is $10 so there is a$240 net loss. But that's not how the family looks at it. The parents just paid $250 for the entertainment of their kids while the kids think that they have made a totally free$10 profit! ;-) Everyone is happy and this kind of "family economics" is of course a template for the "socialist economy" involving adults, too. Maybe it would be a better idea to pay $20 to the kids – who clearly dream about some cash – for vacuuming the house. So I think it's obvious that the relative importance of the – totally inefficient – graphics cards is bound to go up (relatively to the specialized miner hardware) during these collapsing stages of the Bitcoin price. According to electricity, the Bitcoin is becoming even less efficient than ever before. Since January, the hashrate has doubled or tripled – and the fraction achieved by the energy-inefficient graphics cards has probably gone up, too. Meanwhile, many people – early or later investors – want to sell, to have at least something. Most of them are in the red, of course. Most people have invested into the Bitcoin after October 2017 so they simply have to be below zero right now. But the recent deflation of the bubble seems to be increasingly predictable. Most of the people who bought into the Bitcoin did so because of the herd instinct and the "momentum". It went up so they extrapolated the trend and it was always bound to go up, these "clever" people thought. For the same reason, when it's going down, the people start to be capable of seeing that it will probably keep on going down. A self-fulfilling prophesy. This basic dynamics describes the main features of every bubble. First the price goes up, then it reaches the peak, then it goes down, and it can never be revived again. All deviations from his basic profile may be described as rather irrelevant noise. Bitmex has become the most important Bitcoin exchange (although I prefer to watch the trades at Coinbase). It allows both short and long positions as well as leverage. Lots of the long bets have been leveraged. In the recent week, you could see lots of margin calls via Bitmexrekt. During hours when the price is collapsing most quickly, you may see dozens (and sometimes hundreds, per day) of closed long positions worth millions of dollars each. Lots of people are losing lots of money. Even with the natural exponential decrease of the price, the price wouldn't be predicted to go to zero, ever, because the exponential of any real number is positive. So this simple enough model would seem to imply that "the price will never be strictly zero" (zero is "infinitely far on the log scale", we say) although it may go below one dollar again. Well, I think that this model is extremely oversimplified and a more realistic model contains extra effects that allow the price to be strictly zero. The price may be strictly zero because the mining will stop completely and you will be unable to move the coins. There are reasons to think it will be the case. The extra margin calls may help the price declines because the people with the stop loss have to sell their Bitcoin. On top of that, the margin calls may also be ignited by the hedge funds. I think that numerous crypto hedge funds are actually leveraged so they must close their business completely. But even if they're not, they may close because the managers won't get any 2018 salary after the severe drop of the price. If the price drops more dramatically, e.g. below$1,000, then the trading may be intense and the transaction queue may grow again, along with the transaction fees. For a great majority of the Bitcoin holders, even a single transaction fee may exceed their "crypto net worth". So for these people, we might say, the effective value of the Bitcoin will be strictly zero even before the mining stops completely.

It seems rather likely to me – but not quite certain – that we're entering the stage in which not only hedge funds but also cryptocurrency exchanges, mining rigs, and other parts of this ecosystem will start to close down which will further accelerate the selling panic.