According to the efficient market hypothesis, does the price of a stock take into account the company's growth prospects? It says that it is hard to find undervalued stocks but theoretically, is it still possible to get above market returns if the growth is realized? For example, if I invested in Apple 20 years ago, the price may have been fair then but I would still have gotten above average returns from the company's growth. I just want to know what that theory says about this, not whether the theory is correct or not.
The answer to your question is yes, under the efficient market hypothesis, stock prices reflect everything known by traders about the growth prospects of the firm. There are various versions of the EMH but they all say you can't consistently make money using information that is already priced in (such as past prices, known growth prospects, or other public info).
Your confusion seems to be around the difference between an ex ante theory and ex post outcomes. The EMH says today's prices take into account the probabilities of future stock movements. But what will actually happen is unknown until it happens.
The EMH is to financial markets what "you can't beat the house" is to casinos: It's a rule about averages, probabilities, and what can be done systematically, not about individual outcomes. Quite a few people have beaten the house in casinos, but when it happens it is attributed to luck, not skill. Under the EMH, returns above the expected return are good luck and those below are bad luck.
Getting lucky by holding a stock that ends up outperforming is not a contradiction of the EMH unless it was known or could have been known in advance by other traders that it would do so and they did not act on it.
According to the efficient markets hypothesis, does the price of a stock take into account the company's growth prospects?
Yes, it takes in the entire information set available to all actors.
It says that it is hard to find undervalued stocks but theoretically, is it still possible to get above market returns if the growth is realized?
This question is ill-posed. You have to define what you mean by "undervalued." Undervalued in whose opinion? Under what criteria? The theory assumes that the market is in equilibrium. It also doesn't describe what it means by "above market returns."
Imagine a market with two securities. Ex post, one earned four percent and one earned six percent. The arithmetic market average was 5%. One security earned above-market returns, one earned below-market returns. Let us imagine that absolutely everyone on Earth agreed, ex-ante, that the firm that happened to earn the higher realized return was riskier than the one that earned a lower return. Did it still receive an above-market return, * ex-post*? What if it were reversed and the high-risk asset earned a low return, were those statistical errors, were they mistakes of decision-making, or was there a change to the information set ex-post?
Your question needs to be revised to be answerable, however, based on what I think you mean, the realization or non-realization of growth has no impact on outcomes as both possibilities are anticipated and priced and is an example of idiosyncratic risk under the theory.
For example, if I invested in Apple 20 years ago, the price may have been fair then but I would still have gotten above-average returns from the company's growth. I just want to know what that theory says about this, not whether the theory is correct or not. Thanks
This is a dangerous example because Apple was within one quarter of declaring bankruptcy twenty years ago. It was reasonable to presume it would go bankrupt. Microsoft bailed it out in order to protect the IBM architecture from becoming a monopoly architecture. This is a good example of a non-market decision, that is a decision by a single person, Bill Gates, to impact a security.
Had Microsoft not bailed out Apple it is very likely business schools would have drawn the incorrect conclusion that proprietary software systems were so inferior to licensed or open source systems that nobody should ever try the Apple model again.
It would be argued, under the theory, that the extreme discount in pricing was eminently reasonable as a bailout was improbable. In fact, it is only now, with 20-20 hindsight, that it seems like a good idea. For that matter, if you would read the first several years of Amazon's public filings, you will find emphasized and re-emphasized the phrasing that Amazon has never turned a profit, should not be expected to turn a profit and does not have adequate operating cash flows to support its operation. You can also go deep into the history of Mylan Pharmaceuticals and find the same events.
The theory says that people are making consistently rational choices and the valuations reflect the combination of possible realizable paths and the subjective discounting of every actor for each of those paths. The gains are therefore reasonable rewards exchanged for the losses and subpar gains you may have experienced in other securities. As an example, RealPlayer should be totally dominating the music market right now had the most reasonable path that could have happened actually happened. There should be neither iTunes nor Pandora had the most probable path happened. Imagine your losses in betting on the dominant music technology based on the most reasonable path versus the gains on an also-ran hardware firm that had its chance, but messed it up. Why would any rational person invest in Apple?
DISCLOSURE NOTE I am a critic, a really serious critic, of mean-variance finance for mathematical and statistical reasons. I made, I hope, a reasonable attempt to represent the theory as its best proponents would do.
As such, it should be impossible to outperform the overall market through expert stock selection or market timing, and that the only way an investor can possibly obtain higher returns is by chance or by purchasing riskier investments.
In your Apple example, Apple was a risky choice in 1998. Steve Jobs had just returned. He had had some success with NeXT and Pixar, but he was not an obvious management genius at that point. Products like the iPod and iPhone weren't even on the radar then. It was to some extent luck that things worked out the way that they did.
Other examples include oil surveying and pharmaceutical development. Two companies in the same field could have identical characteristics, but one might be developing a drug with fatal side effects while another might have lucked into Viagra. One company goes broke, while the other becomes fantastically successful. If you happened to pick the right company, great. But the efficient market hypothesis says that you are just as likely to pick wrong. So it recommends investing in both or neither. You won't beat the market that way, but you can match it without risking lagging it.
This does not mean that you can't make above market returns. You can. But the strategy to do that is to take risks that pay off. When you take risks, you can also get below market returns. So the efficient market hypothesis says that for every Warren Buffett, there's some number of other people who took similar risks and went bankrupt.
This is of course contestable. But you wanted to know what the hypothesis said rather than establish its truth.
The efficient markets hypothesis states that every piece of data is priced into the stock. But what the hypothesis does not take into account is market sentiment - people's fear and greed. It also doesn't take into account that valuations are based on assumptions, assumptions based on the bias of the analyst (that is why you get different valuations for the same stock).
So of course you can get above market returns. There are many stocks that perform better than average market returns and there are many stocks that perform worse than market returns.
Using a simple strategy of only investing in up-trending stocks when they are up-trending and to get out of them when they start down-trending, you can easily beat the market.