Stock Market Efficiency

I've been thinking a lot about stock market efficiency lately. When I say "efficiency" I mean the ability of the stock market to factor in information relevant to the price of stocks very quickly. One related aphorism is, "Buy the rumor, sell the news." In other words, well before any announcement is made by a company, the stock price will have increased based on the diffusion of that coming information through society. This leads to some interesting questions. Consider a hypothetical conversation that I'm sure has taken place innumerable times in American history.

Alice: I have some money saved up. What should I invest in?
Bob: Stock market. Index fund.
Alice: Why?
Bob: Because the stock market has historically returned 10% per year for over a hundred years.

Clearly, there's something wrong with Bob's statement. There's nothing factually wrong with his statement; the stock market has indeed return around 10% per year. Rather, what's wrong is Bob's implication: that the stock market will keep returning 10% a year into the future.

If the stock market is truly efficient, those expected returns should be factored into prices already. There should be a premium in prices that takes into account the expected 10% returns. Am I correct?

Share this

I think the EMH is wrong but

I think the EMH is wrong but I don't think this particular complaint is fatal. Another tenet of modern finance (which I also think is wrong) is that average returns are correlated with risk or beta. It's quite clear that the stock market has long periods where it generates poor returns (like the last decade) and many investors need less volatility for their savings than the stock market provides.

I believe the standard answer to your criticism would be that stock market investors receive a greater average return in exchange for taking on more risk/volatility. I don't believe that is correct but I think you need to make a more nuanced argument to explain why the standard answer is wrong.

Fallacy of Division

What is true of stocks as a whole is not necessarily true of any particular stock. Therefore the expectation that "stocks" will return X% does not feed back into the pricing of any one stock.

I'm skeptical

Your answer implies one can "bypass" what should be a built-in price premium (for expected returns on stocks) by investing in an index fund.

If the stock market is truly

If the stock market is truly efficient, those expected returns should be factored into prices already. There should be a premium in prices that takes into account the expected 10% returns. Am I correct?

Why not? So that premium could well be factored in already. I hope you're not trying to say that there should be a 0% return once the premium is factored in. You're not forgetting time preference, right?

But time preference can't explain the 10% return

That is indeed what I am implying: that there should be a 0% return once the premium is factored in.

If your explanation is time preference, then why is the return on stocks so much higher than the return on keeping money in the bank? (I realize that now we're getting into one of the fundamental mysteries of economics--the equity premium--and fully acknowledge that I lack any insight on the matter).

That is indeed what I am

That is indeed what I am implying: that there should be a 0% return once the premium is factored in.

Then you'll have to explain why you think that. Your next question implies that you accept that it should at least give the same return as keeping money in the bank, and that is not 0%, so I'm confused by your simultaneously thinking that it should be 0%. I'm also not sure if you are rejecting time preference.

If your explanation is time preference, then why is the return on stocks so much higher than the return on keeping money in the bank?

How should I know? I'm not an economist. But since you ask, as a non-professional it's what I would expect. Banking is warped beyond recognition by the state. It would take me months to really understand it so I don't and I can only reason by analogy to other things warped beyond recognition by the state, and one common element is that benefits are low and prices are high. The state monopoly takes its cut. But if you put money into a bank and then take that money back out later on, the only place for this cut to come out of is the money you get back. Hence a low return on keeping money in the bank.

In contrast, buying and selling stocks is decentralized. If you buy a stock, might be buying it from me. The middle-men of course take their cuts and those cuts are probably larger than they would be in a freer market, but we know what those cuts are and they're pretty small (the fee you pay to buy stock).

Let me revise

Then you'll have to explain why you think that. Your next question implies that you accept that it should at least give the same return as keeping money in the bank, and that is not 0%, so I'm confused by your simultaneously thinking that it should be 0%. I'm also not sure if you are rejecting time preference.

Let me revise: The returns shouldn't be zero. They should approximate the "market rate of time preference". At the very least, they should be far less than 10%.

Do we know what is the

Do we know what is the market rate of time preference? Do we really know that it is more closely measured by one thing (e.g. bank accounts) than by another (e.g. stocks)?

Anyway, checking out the Wikipedia article, it's got enough there to spend a long time evaluating, but one thing stood out:

Empirically, over the past 40 years (1969–2009), there has been no significant equity premium in (US) stocks.

I remember back when cold fusion seemed to be discovered. A lot of people started explaining it. Then it turned out to be irreproducible.

equity premium

You are asking for the answer to the equity premium puzzle - nothing more, nothing less. You mentioned it briefly in another comment, in a way which seemed to imply you thought you could avoid the equity premium puzzle while asking what the answer to the equity premium puzzle is, which is pretty silly!

The high variance of stocks

The high variance of stocks is one reason. It's related to time preference - not everyone is equally willing or able to risk their investments going through prolonged periods of below risk-free returns. If your money is in the bank then you can withdraw it at any time to cover unexpected events (job loss, health problems, other unexpected life changes) without losing money. Investments in an index fund may go through long periods where cashing them in would result in a loss, even if the expected return is greater over a long enough time period. This is also why the standard advice is to shift your money out of stocks as you near your expected retirement age.

Some investors need lower variance in their returns and are willing to pay a 'premium' in the form of lower average expected returns to get it. I don't think that's the full explanation for the higher expected returns of stocks but on it's own it is enough to explain some of the equity premium.

The problem is believing the

The problem is believing the stock market holds with your definition of "truly efficient".

When people talk about efficient markets and the stock market, they are generally arguing that it isn't possible to beat the general market via anything other than luck. This efficient market definition is all that is needed to tell someone to invest in an index fund over their own stock picks or a mutual fund that has done well recently, so there isn't a contradiction like you were imagining. Your expanded definition of "discounting all possible future gains" doesn't need to hold for index funds to make sense for an average investor.

Of course, whether or not the 10% returns that people have relied on are expected to continue in the future is another story as you mentioned in your above comment, but you aren't going to get that answer by imagining rules for a magical hyper-efficient stock market.