The big news is that Bitcoin has now hit a new record, surpassing the $4,700 mark. As my regular blog readers (and podcast listeners) will know, I have some reservations about the current wave of apparent crypto-enthusiasm. Don’t get me wrong – I like the idea of bitcoin and cryptocurrency in general. And I’m sure that crypto is going to be the future of money.
Because, not to be facetious, but how could any central bank (or tax authority!) resist? Blockchain technology offers the allure of a massive public ledger, where every single transaction is broadcast to (and stored by) every computer on the network. Imagine a world where every asset and/or every casual purchase is a matter of public record. It’s triple entry accounting – every cryptoRand (or cryptoDollar) would contain both its current ownership, and all of its ownership history. All you’d need is the right algorithm to read it.
But that dystopian vision aside, that’s not where all my bitcoin-specific reservations come from.
So let me ask some questions.
Question 1: What Is Really Driving The Bitcoin Price Surge?
I know that this is the million dollar question. But all the standard explanations don’t really seem to check out.
So before anyone tells me that the price surge is just a sign of bitcoin achieving mass acceptance, here is a chart of bitcoin trading volumes for the last two years:
Just to be clear: the total volume of bitcoins traded has collapsed this year. And it was after the collapse that the price surge happened:
This is not normal. When you get price surges, you expect volume surges. Not this.
But perhaps you’re not persuaded by that. Perhaps you’re thinking that the volume anomaly is just a reflection of the price going up – so people can’t buy as many ‘whole’ bitcoins. Maybe they’re buying parts of bitcoins.
Well, let’s look at the number of trades per minute for the last 2 years:
Well that’s also near historic lows. And the dramatically lower number of trades per minute happens to coincide with the period of the price surge.
Again, this is not normal. You wouldn’t expect a drop in trading activity to correlate with massive price surges.
So what the hell is going on?
Here’s a possible response that I’ve heard before: “Oh, the data is wrong – the figures coming out of China are notoriously unreliable.”
Question 2: is there fake data?
Well, here’s the chart of trades per minute, split by currency:
So whatever else, the yuan (CNY) trade appears to have dried up in 2017. Was 2016 just an anomaly for ‘fake’ bitcoin trade data (which we can now disregard, because the yuan trade has disappeared)?
Unfortunately, I don’t know.
Although if I’m honest, I think it’s weird that all this fake data would suddenly ‘correct’ itself right as the bitcoin network started having problems processing transactions. Just to remind you, the network first started having serious issues with processing times toward the end of 2016:
Isn’t this a more plausible reason for the drying up of the yuan trade? If trades stopped happening as quickly because the network couldn’t take the pressure – then that would certainly scare me off (if I were a Chinese day-trader).
But that aside, let’s just look at USD trading volume data alone then:
Admittedly, I’m looking at the volume of USD-BTC trading volumes denominated in bitcoins. But it makes more sense (to me, at least) to look at it this way. Because there’s no discernible pattern here that says “Look at people buying lots more bitcoin.”
And remember: this is assuming that all the Chinese yuan trading data is false.
If you assume that the trading data is mostly accurate, then you’re back to being baffled by really low volumes of trade.
And if you assume that the data is mostly false, then why are we even here?
Question 3: how is the BTC:USD price measured anyway?
I realise this sounds like a semantic question.
Here’s the thing though: Bitcoin (or BTC) trades on multiple exchanges. But when I look up the USD price of BTC on Google or Coinbase, I get a single price.
And I don’t know where it comes from.
Is it a straight average of the prices on all the exchanges? Is it a weighted average? Does it prioritise some exchanges and leave out others?
This is not a minor question.
For example:
- Let’s say that there are only two bitcoin exchanges.
- The first bitcoin exchange is where everyone trades, and the last traded price was for $4,000.
- The second is barely used, and seems to have some shady questions around where it came from. It mainly seems to be a place to launder money (because Bitcoin exchanges are unregulated, this is entirely possible). And the last trade took place at $10,000 (some heavy laundering, right there).
- If Google and Coinbase just use a straight average, then the quoted BTC:USD price suddenly soars to $7,000.
- On the other hand, if they use a weighted average of all trades closed in the last 24 hours, the BTC: USD price would barely register the $10,000 anomaly – and the price would remain constant at $4,000.
So the methodology is important.
Only, I can’t tell how anyone has arrived at the ‘average’ Bitcoin price.
My main point
The bitcoin network is meant to be decentralised, transparent, and self-regulating.
Right now, it looks opaque.
And if this were any other asset class, I’d be saying that the entire market looks like it’s being manipulated.
Rolling Alpha posts opinions on finance, economics, and sometimes things that are only loosely related. Follow me on Twitter @RollingAlpha, and on Facebook at www.facebook.com/rollingalpha. Also, check out the RA podcast on iTunes: The Story of Money.
Comments
Anonymous August 30, 2017 at 10:18
Its a classic pump and dump…one for the history books someday
ReplyMark August 30, 2017 at 12:55
i think in part it is being in the same way as the Cape Town property market.
finite supply — things going badly elsewhere (safe haven) — get rich quick gen X mentality — media hype
the other reason is that more countries are like South Africa then are like Germany, ie they have populations that are extremely sceptical of the governments, they are restricted by currency exchange laws, tax laws etc and rightly or wrongly they think that their currency is “under performing” Bitcoin offers an extremely attract alternative to this kind of person/market
ReplyJayson August 30, 2017 at 13:30
Hey Mark! My problem with this kind of explanation is: if that were true, then we’d see different trading volume data. I don’t think we’d see a massive drop in trading volumes. Yes, the supply may be finite – but the volume of trade in that finite supply should be on the increase!
ReplyJustin August 30, 2017 at 15:15
Hey Jayson. Any correlation between this volume slump and the increased volumes traded of Etherium/Dash/others? Perhaps Bitcoin valuation has put it beyond the grasp of the speculators wanting to make a few bucks who may have now moved to cheaper currencies (ignoring that you can buy fractions of coins :P)? Also, most press seems to be about the “Bitcoin Bubble” – there is less said about other cryptos so perhaps less perceived risk?
ReplyJayson August 31, 2017 at 07:14
So I wanted to share a comment that came through Facebook, because I think it’s important (especially the part about my data attribution).
Here’s the original comment:
“Look, I’m no economist, so I’m probably out of my depth having this discussion, but your charts seem to be at odds with every other trading volume chart I can find. Where did you source yours from? http://www.zerohedge.com/…/daily-trading-volume-bitcoin…
“Also, the volume chart at the bottom of the Bitstamp real-time chart shows massive spikes (green and red respectively) when there are surges and dramatic dips in price, which is entirely consistent. So, I’m battling to understand where your data comes from.
“Also, on the issue of exchanges, it seems like a bit of a straw man argument: while some undoubtedly do a lot more volume than others, I watch Bittrex, Luno, Bitstamp and SpaceBTC, and they’re all within around $20 of each other at any given moment. I think the crypto community would red-flag exchanges that seemed to be dramatically out-of-kilter. In fact, that’s something you don’t seem to acknowledge here: the power of a kind of ‘real-time peer-review’ through the online communities of Reddit, Steemit, YouTube and Twitter, to verify trends and highlight scams.”
And here’s my response (most of it – I’m leaving out the greetings, etc):
“The chart from Zero Hedge seems to show the trading volumes for only the USD:BTC trading pair – the problem is that other trading pairs (like the CNY:BTC) drove much (much!) more of the bitcoin trades prior to 2017. The volume trading chart that I used showed all currency trading pairs (and you can play with it on data.bitcoinity.org).
“The other issue is that even where Zero Hedge and I are talking about the same chart (ie. the BTC:USD trading pair), I’m denominating the chart in Bitcoin, where Zero Hedge is denominating the chart in US dollars. I did mention this quite specifically in my post – but perhaps I wasn’t very clear about why I’m using the bitcoin denomination. When the volume chart is denominated in US dollars, you’re incorporating the price data into the volume data (so the higher prices would make it look as though the trading volumes are increasing). But my question is about the bitcoin demand itself – that is, I’m questioning why the number of bitcoins being traded seems to be falling. And for this reason, I’m denominating the chart in bitcoins (and stripping price out of it). Basically, my argument is: “If there is a higher demand for bitcoin due to mass acceptance, you should see more bitcoin trading hands. Only, this isn’t happening.”
“Finally, regarding “real time peer review”, I don’t mention it because I don’t really believe that it works. Technically, the stock market gets real time peer review from market participants, but you still get mis-pricing. Unfortunately, history is full of pricing bubbles – in which people prove that they’re really not good at self-reviewing their own investments. I’m not sure why crypto would be any different (if anything, it might be worse – because crypto investment is often accompanied by a particular set of political beliefs about the world). But perhaps that’s just my cynical side coming through!
“Either way, definitely check out bitcoinity.org.”
ReplyTimothy Van Blerck August 31, 2017 at 22:02
I think there may be an issue with the data you are pulling here
1. The faster the network is processing transaction the smaller the mempool size, hopefully once segwit improvements kick in over the year this should improve.
https://blockchain.info/charts/mempool-size?timespan=1year&daysAverageString=7
2. Confirmed transaction per day does show a drop in June 2017 but not as large as shown in your graphs. Also August is picking up now that Segwit lock in decision has been implemented stabilising confidence.
https://blockchain.info/charts/n-transactions?timespan=2years&daysAverageString=7\
3. USD exchange trade volume is steadily growing with a new high this month
Replyhttps://blockchain.info/charts/trade-volume?timespan=all&daysAverageString=7
John McLaren (@indiefutures) September 14, 2017 at 10:46
Let me start with your main point where you say “The bitcoin network is meant to be decentralized, transparent, and self-regulating. Right now, it looks opaque. And if this were any other asset class, I’d be saying that the entire market looks like it’s being manipulated.”
I think you’re confusing the Bitcoin network with the market and the exchanges. The transparency, self-regulation and decentralization stems from the immutability of the cryptography and the difficulty of the proof-of-work. It’s the disintermediation of the trust element that makes transactions transparent. However, that does not prevent anyone from trying to manipulate the short-term price for short-term gain. As long as markets have been around there have been attempts to manipulate them.
I can offer my own reason for the volume question around Bitcoin that you have. All the dirty tricks that have been criminalized in regulated markets are at play in this unregulated market. I suspect that the two main culprits are dark-pools and quote spoofing by high-frequency algorithms. Originally, dark pools were created as a vehicle for institutions to conduct large block trades with one another to mitigate the adverse effects on price associated with trading large blocks. Even an extremely liquid stock, when met with a large one-sided order, can be moved significantly. Dark pools were designed to match one party with an unbalanced trade to another party. This matchmaking minimizes the impact of block trades and thereby reduces volatility of the markets. In regulated markets dark pool pricing is based off the consolidated tape, thereby preventing trades from occurring outside the bounds of the best bid and offer. This works so long as the vast majority of the market depth is behind that of the consolidated tape. As the market share of dark pools grows, it essentially drives away the depth and volume behind that of the consolidate tape. This causes the information behind the public’s consolidated tape to become more and more erroneous as the private privileged market volume avoids displaying its quotes to the consolidated tape. As liquidity is siphoned off the exchange, the information behind the public’s consolidated tape becomes more and more erroneous as the private privileged market volume avoids displaying its quotes to the consolidated tape.
I think the vast majority of volume right now is in dark pools by entities not wanting to show their hand and as there is no official tape or bid/ask price it’s wild west territory as far as trading is concerned.
Secondly, high-frequency traders are definitely spoofing the different exchanges in different countries for arbitrage opportunities. They provide fake liquidity by cancelling most of their orders after a few seconds in an attempt to manipulate the bid/ask price. These actions create a false impression of buying and selling pressure. The appearance of these ingenuous orders may cause other market participants to readjust or hesitate. So yes, the market is being manipulated but that has nothing to do with the transparency of Bitcoin’s blockchain. If you can find the volume of cancelled orders or the order cancellation ratio as well as the quantity held in dark pools you may have the answer to your question.
I think short-term attempts at understanding Bitcoin pricing are fruitless. Satoshi Nakamoto designed the protocol so that there would be a slow, steady increase in price over time. The cryptography underwrites this. He set it in an ecosystem based on Austrian Economics which would guarantee it’s success and long-term viability. Genius. What we are witnessing with Bitcoin is the intersection of an Austrian ecosystem with a flawed market run on Keynesian principles. The better economic model will win.
ReplyJohn McLaren (@indiefutures) September 14, 2017 at 14:51
Another point to look at is the volatility of BTC. Whilst it is not yet included in the VIX you can get an idea by looking at the 30-day exchange trailing average. Is the trend up or down?
An interesting piece of data I picked up on is that the top 500 richest Bitcoin addresses hold 40% of all Bitcoins in circulation. So liquidity is a key issue in determining price fluctuations and not necessarily volume per se.
ReplyJohn McLaren (@indiefutures) September 16, 2017 at 00:08
The way I analyse volume is like this :
ReplyKeep a running total of the daily volume, adding the volume of a price bar when it closes higher than the prior, and subtracting the volume of price bars that close lower than the prior. If the mean of this number is falling while price is rising it means that the trend could be about to reverse. The opposite holds true as well.
Cius October 27, 2017 at 08:57
The current cost of validating one bitcoin is now 203 KWh of electricity. That is the energy consumption of a small home for an entire month. In other words if you buy a R50 sandwich with bitcoin somewhere in the world some miner is paying about R300 in electricity costs to validate the transaction. He is paid in more crypto currency (essentially printed money). This is the ponzi scheme nature of the beast. Unlike with fiat money where the transaction fees are built into the cost of an item when you swipe a credit card the transaction fees for bitcoin, which is mainly electricity, is paid for by a third party who gets rewarded in more crypto currency. It only stays alive while the miner is able to sell his new coins to new entrants to pay his power bill and take his profit.
There are several hard limits to this that will cause it to implode. Firstly getting new power stations online takes forever and is very expensive. Bitcoin already uses as much power as Ecuador. And this is with bitcoin being a tiny percentage of the volume of fiat currency. There is simply not enough power in the world to switch to crypto currency at any kind of large scale.
The second limit is the normal ponzi scheme limit. When new people are unwilling to buy the new coins being issued the miners will very quickly turn off their mining rigs (as it costs them real money to pay the electricity bill). At that point you can’t validate transactions and it all comes tumbling down.
The scariest aspect for me is the environmental aspect of what this is costing! The pollution caused by crypto currency mining is astronomical.
I think we should mature the blockchain technology in other areas that make more sense such as elections, asset ownership, etc. Maybe at some later point it can move to currency. However when we get to that point we will have to have solved the problem of transaction cost and somehow included the cost as part of the buyers transaction costs to avoid printing money. We also will have to find a way to limit blockchain size to stop the transaction fee increasing endlessly.
Reply