This piece on the efficient market hypothesis (EMH) by Nick Carter that explains the hypothesis as well as anyone I’ve come across. I agree with everything Nick says up to the point where applies the model (as he rightly describes it) to the question of whether or not the halving of bitcoin supply has been priced in or not.
Nick, in the first half of the essay, takes us through the history, definition, and a simple explanation of the EMH. He goes through great pains to explain the model’s limitations as well, but in my opinion, he doesn’t go far enough.
The EMH has no predictive power because it can only say what is priced in today, not what will be priced in tomorrow. True, the EMH states that the expectations of participants (i.e. what they believe will be priced in tomorrow) is priced into the current market price of the thing being priced, but it doesn’t say whether those expectations are correct or not. So yes, obviously the halving of bitcoin is priced in. But the much more interesting question is whether the halving is priced in correctly?
The idea that the halving is already priced in (correctly) because the supply schedule of bitcoin is out in the open and in the code, easily verified by anyone willing to do so, is a mistake of first order thinking. Second order effects of the halving like press attention, more awareness, and more interest in the supply schedule of bitcoin (as one example of second order effects) will have a massive impact on the fundamentals of bitcoin adoption. And these second order effects are impossible to price in correctly. Market participants will get it wrong, whether they are overly optimistic or not optimistic enough is the main question and, in this regard, EMH offers literally no help whatsoever.
It is because of this that I disagree with the second part of Nick’s essay where he uses the EMH to explain that the halving is already priced in. Not because I don’t agree that the halving is priced in or not (it most certainly is), but because he uses the EMH as if it tells us anything about the future price movements of bitcoin (which it most certainly does not). It’s not that Nick is wrong, it’s just that he’s not right either. Although EMH is a great explanation of how markets work, it offers very little practical value with regards to what is actually priced in and to what degree (future price movements).
What’s priced in today will change tomorrow as new information and new participants enter the market. This is the adaptive markets Nick mentions. But because of this adaptive nature of markets, what is priced in today only describes what is priced in today and not what will be priced in tomorrow (except expectations of participants of what will be priced in tomorrow). In essence, EMH is a static explanation of a very dynamic system and it should never be used as a method to explain future price movements.
An example of second order effects of the bitcoin halving:
This is the clearest example I can think of (by no means the only one) which shows the second order effects of the bitcoin halving. As we come closer to the halving the noise around it increases and the awareness of bitcoin itself increases. I for one know that because of the halving I have had conversations about bitcoin with pre-coiners that might never have happened. I saw the first ever article covering the halving this morning in one of South Africa’s mainstream newspapers.
But the chart above is already priced in (that I agree with fully), either through new participants entering the market or other participants becoming more confident that more people are becoming aware of bitcoin’s monetary policy. What is not priced in is what this chart will look like in say, two months.
Imagine that chart spikes over the coming months, what would happen to the price? It depends. Did market participants think it would spike more or less? If less, then the price they are willing to pay will drop in their estimation. If more, the price would increase in their estimation.
The EMH tells you absolutely nothing about what that chart will look like in two months, only that in two months the new price will price in the new information.
Bitcoin market is not efficient enough to price in halving correctly:
As someone who spends a lot of their time trying to accurately price publicly listed entities, I can tell you with near certainty that bitcoin is not close to being as efficient a market as even some of the most obscure small caps the world over. To explain this, I will use an example of Netflix vs. Bitcoin for price setting.
These two have somewhat similar problems for market participants looking to price them. They both are new technologies we haven’t really seen before; they are truly global; we have no idea what the size of the market is they will capture; valuation is dependent on adoption (subscribers for Netflix and users for Bitcoin).
They are also different in ways that makes for good explaining as to why the market for one of them is much more efficient than the other.
Beginning with Netflix let’s look at some of the necessities for an efficient market:
Free market – Netflix is listed on the NASDAQ in one of the freest countries on the globe with some of the most market friendly regulations one will find. As an outsider that’s looked at a few markets, there is no doubt that the US has one of the best regulated stock markets in the world. Some regulations do prohibit some information to float freely, but I contest that this is miniscule in the overall picture (all insider information eventually comes out as soon as it becomes relevant).
Sufficient liquidity – Netflix has a market cap that is (serendipitously for yours truly) very close to that of Bitcoin of around $150b dollars. That’s nice liquidity which definitely helps to increase the efficiency of the market. But looking at the market cap of a single stock is wrong. Netflix forms part of a much larger market where participants are looking for miss-priced opportunities. The total US stock market value is around $30T. That makes the almost $200b cryptocurrency market look like nothing (which it still is). This liquidity makes for more participants hunting and thus a more efficient market. Because of this liquidity of the total market (along with the other requirements being sufficiently fulfilled) I will contend that even some of the most obscure small caps are more accurately priced than bitcoin.
Easily accessible fundamental information – Netflix has so many reports (10Q’s, 10K’s) that it is difficult to keep up with. These reports are designed to give market participants the most relevant information for the pricing of the stock of the company (admittedly they aren’t perfect). There exists an abundance of sell-side analysts whose only job is to source more relevant information to help market participants price these stocks correctly. Management give guidance of their expectations on things like growth in subscribers, ARPU, content costs etc. etc. all the most relevant information for the accurate pricing of the stock. Fundamental information is abundant and easily accessible to almost any participant to varying degrees.
Sophisticated participants – The stock market has existed for more than a hundred years, it has massive liquidity, and it is extremely ruthless on people who get things wrong. Thus, only the most sophisticated of participants are left to compete for the scraps of what is now very few miss-priced opportunities.
Valuation (or ability to price accurately) – This is the most obvious and important disagreement with the idea that bitcoin is even close to being an efficient market. Our ability to price anything accurately requires a framework of how to value said product, stock or commodity. Valuing Netflix, although very difficult, is a lot easier than valuing bitcoin. We have things like Discounted Cash Flows (DCF), Price-to-earnings, Price-to-book and a score load of other valuation methods that forms part of the analyst’s framework and any new information is easily digested through this framework and incorporated in the price. In short Netflix’s value is driven by subscribers, revenue per subscriber, and costs to get subscribers; throw your assumptions (which are heavily guided by management) into your DCF and you come close to what you believe the share price will be. Easy….ier than valuing bitcoin.
Now turning to bitcoin:
Free market – The freest the world has ever seen (Thank you Satoshi!). Although there are some regulatory hurdles that make it less free especially on the on-ramp side of things. Another factor that decreases access to this market is regulations for asset managers which prevents them from even considering bitcoin as an investment.
Sufficient liquidity – At a market cap of $150b bitcoin is by no means small but looked at as a complete market, it’s not even close to attracting the most sophisticated investors yet.
Easily accessible fundamental information – One of the best things about bitcoin is that every bitcoiner wants every person to know all the news around bitcoin. The incentives are for participants to share all (most of) the most relevant information they can find (Thank you Satoshi!). But even with these incentives, good, reliable, and relevant information around fundamentals are still pretty hard to come by. We’re not even sure about how many people currently own bitcoin, never mind other more obscure fundamental information. Great news for any person trying to take advantage of miss-priced opportunities.
Sophisticated participants – Although the sophistication of market participants has increased significantly over the lifetime of bitcoin there is still a lot of what is endearingly called “dumb-money” in this space. It’s the retail first revolution, and that makes for some very unsophisticated participants indeed! In short there are a lot of noise-traders willing to sell you bitcoin when it’s the right time to buy, and lots of bag-holders willing to buy from you in December 2017 (-21?).
Valuation – Can someone please give me ONE good way to value bitcoin? And if you say Stock-to-flow I will punch you in the fucking face. There is no decent way, in my opinion, to value bitcoin as of yet. I don’t know if there ever will be – just look at gold. Now, without a framework of how to value the thing we are trying to price, how will we ever be able to price new information correctly? This is a massive barrier to market efficiency in my opinion, especially in this very early stage of bitcoin. Most people investing in bitcoin are thinking in terms of portfolio allocation when allocating capital to this market, not whether the asset is under- or over-valued. As this asset grows so will the valuation framework around it, but so far, we haven’t come close to anything concrete.
Based on these above factors, and there probably many more I have ignored for the sake of brevity (I have a day job and I’m not allowed to work on things we can’t invest in), I think it’s pretty easy to say that the market for bitcoin is nowhere close to being efficient yet. My bet is that it will become much more efficient over the coming years, but that’s already priced in.
I have no gripe with the EMH being used to explain how markets work, I think it actually does an amazing job of explaining it. I think every bitcoiner, or any market participant in any market for that matter, should understand the EMH and its implications. But it is also very important to understand its limitations.
EMH can, by definition, not be used to predict what future price movements will be. All it can tell you is that as of today all available information is priced in. No more and no less.
So, is the halving priced in? Absolutely.
Is it priced in correctly? Absolutely not.
Is this good for bitcoin? Yes.