the hypothesis maintains that
stock prices reflect all relevant
information about the stock
This is a common description of the EMH. But every time I read it, I think: Does information really directly impact the price of a stock? How?
What if it takes 12 months of hard thinking to draw the right conclusion from the information? Are there many investors who go to such lengths? Are they all thinking at the same speed? And if not, what does that tell us about the EMH?
Google released DeepDream in 2015. My feeling is that with enough thinking, one could have predicted where image generation is going in the next decade and that language generation would go a similar route. And that this will lead to a high demand in Nvidia's GPUs. But that thinking would not be instantly. It would take months or years.
Information that requires 12 months to figure out isn't information that's available now.
Say you want to know the 400 trillionth digit of pi. We have all the information needed right now to know how to compute it. But you don't know what the actual digit is yet. The information isn't available and won't be until you set your supercomputer on it for some number of months. Having the information necessary to derive other information isn't the same as having the derived information.
If there is some information about a future stock price that could theoretically be computed after months of work, that's still not information that currently exists, and therefore is not currently reflected in the price. If no investors go to the lengths to get that information, it'll continue to not affect the stock price. It's not violating EMH because it's not information that exists yet.
That definition would mean that smarter investors, who can think faster and further ahead, get information faster. And therefore have information now that others do not.
That seems to be directly the opposite of the common definition of the EMH, which emphasizes how the market reacts to new information. And not how it produces information. For example in TFA:
"the market rapidly responds to new information"
Wikipedia starts the "Theoretical background" with an example on how information becomes widely available to all investors, not how one fast smart thinker generates it:
Suppose that a piece of information about the value
of a stock (say, about a future merger) is widely
available to investors.
The smartest, fastest investors are the ones who make a profit by incorporating their information into the stock price in the EMH. The stock price can't move on its own. Under the EMH, someone has to be the first to trade stock based on information so that the stock price reflects it. When they say "the market rapidly responds to new information", that means investors with the new information are buying or selling accordingly. It's not opposite at all.
How the information gets produced is irrelevant to the EMH. Whether it's obvious or takes hard thinking, either way, once investors obtain the information, they will trade based on it, and that will move the stock price.
>That definition would mean that smarter investors, who can think faster and further ahead, get information faster. And therefore have information now that others do not.
> What if it takes 12 months of hard thinking to draw the right conclusion from the information? Are there many investors who go to such lengths?
It's not required to be all of them. Suppose that it indeed isn't, but the ones who do that work for investment funds who control significant pools of money.
Now the investors in two or three of those places do the research and conclude that some company is about to start doing well and their share price is currently $50 but is about to be $150. So they start buying it, and keep buying it until it gets up near $150. Which happens pretty quickly because they control enough money to use up all of the short-term liquidity at the lower prices and the majority of the shares are held by people who aren't even paying attention and therefore don't try to sell when the price starts going up. Once the price gets to that point they don't buy any more because it's no longer selling at a discount.
Then the company actually starts doing well to the point that everyone can see it but the price hardly moves because it was already priced in.
That would mean that the p/e-ratio of a company would rise sharply long before the profits set in. And that rise would be called "mysterious" by the general public. And then only when the profits set in, the p/e would come down.
The landmark paper, "Attention is all you need", that triggered the breakthrough that led to current transformer architecture LLMs, only came out in 2017. Without that breakthrough, they wouldn't exist. And even then, the early models produced gibberish. Better gibberish than older Markov chain text generators, but asking GPT-2 "What is three plus five?" would give some nonsense, non-sequitur answer, that might start with a (incorrect) number if you were lucky. At the time, everyone was wondering if scaling up the model size would improve intelligence or hit a wall. ChatGPT didn't release until 2022.
And you'd need to know back in 2015 that Nvidia specifically would be the big winner from AI. They don't even manufacture their own chips. Intel also designs chips and GPUs, but if you bet on them in 2015, you'd have lost money between then and 2025.
> That would mean that the p/e-ratio of a company would rise sharply long before the profits set in. And that rise would be called "mysterious" by the general public. And then only when the profits set in, the p/e would come down.
You have to look at the volumes involved: if there are tens of millions of shares of a particular stock moved everyday, a single event that involves 100,000 shares is going to be lost in the noise.
There are always people who think they know better (if they didn't think so they wouldn't be trading), and they may make crazy-appearing trades. Lots of the people in The Big Short were viewed as 'lunatics' ("You're betting against the housing market?") that turned out to be right. But also remember that there are people who think the world is flat.
> The price roughly rose along the earnings. Even though the foundations for generative AI became clear in 2015.
It's also why you hear the talking heads on television say things like "…this has already been priced in.".
You're not likely to see that in huge companies because everybody is already paying attention to them and it's harder to know something someone else doesn't about the thing everybody already knows everything about. Also, then it's more likely to happen on a scale of 10 days than 10 years.
Where that really happens is with startups and younger companies. Some company is currently making negative dollars but a few people have figured out that they're likely to be big so their share price is up before their earnings are.
And suppose you somehow actually knew what every major company's earnings would look like in every year from 2015 to now. Do you invest in Nvidia in 2015? Or do you invest in Netflix in 2015 and Tesla in 2019 and so on and not bother with Nvidia until just before the hockey stick?
> What if it takes 12 months of hard thinking to draw the right conclusion from the information?
I think the idea behind EMH is that this probability is priced in, at any point in time. It just so happens that longer term probabilities are discounted as more volatile, thus impacting less the present price.
In systems thinking there’s the concept of “stocks” or “buffers”. Meaning that change of inputs into the systems first affect stocks/buffers before the outputs.
Information asymmetry is a thing, but EMH nominally handles that; prices quickly shoot up to the max any one guy is willing to pay
Suppose we have a stock and a bunch of investors. All of the investors have some set of information implying a value v. Except, one investor is smarter and finds an edge (rise of AI or whatever) which implies a value of 2v.
That investor will buy the stock any price up to 2v. The rest of the investors will be happy to sell at any price above v. Given unlimited money the price should stabilise at 2v very quickly.
However there are lots of real-world caveats like, not everyone has an infinite money glitch.. and there are probably second-order and third-order effects like some hedge fund notices the pattern and does XYZ which influences price... options make price a function of expectation of price, then the price of options is driven by the expected price of the same options near execution date... idk
you're wrong about the mechanism - it's not that the thinking is the cause of the efficiency. It's the large number of participants all doing their own brand of thinking, and that the _average_ of all of those approaches the "correct" price. It requires the large number of participants because for such an average to approach "correct", errors within each participant's guesses cancel each other out.
And the immediacy comes from the large amount and speed of the transactions. It does not require that these participants sus out the correct value from information - they could've actually just guessed.
> What if it takes 12 months of hard thinking to draw the right conclusion from the information? Are there many investors who go to such lengths? Are they all thinking at the same speed? And if not, what does that tell us about the EMH?
To paraphrase William Gibson: the information may be available, but it is not evenly distributed.
It's why (e.g.) hedge funds use satellites to get information on company activities:
It's takes resources (time, money, etc) to gain an advantage, and it's only do it because they think some extra bits of information will allow them to know more than The Market in general / their counterparties to get a better conditions on a trade or options.
Why do you think insider trading became illegal: some folks have that information before others simply because of their job/position. There was a case of someone knowing something early, because information can only travel as fast of the speed of light, which some "beat":
> Last Wednesday, the Federal Reserve announced it would not be tapering its bond buying program at 2 p.m. ET. The news takes seven milliseconds — about the speed of light — to reach Chicago. But before the seven milliseconds was up, a few huge orders based on the Fed's decision were placed on Chicago exchanges.
EMH is saying people that if people think they can make money, they will spend the resources to get an information edge to accurate price what a commodity is 'worth', either higher or lower. If you better know what it 'should' be, then you can devise a trading strategy (buy/sell/short/long) to get one over your counterparty.
What if it takes 12 months of hard thinking to draw the right conclusion from the information? Are there many investors who go to such lengths? Are they all thinking at the same speed? And if not, what does that tell us about the EMH?
Google released DeepDream in 2015. My feeling is that with enough thinking, one could have predicted where image generation is going in the next decade and that language generation would go a similar route. And that this will lead to a high demand in Nvidia's GPUs. But that thinking would not be instantly. It would take months or years.