I like it overall but as any analogy, it must be used with caution as it fails to infer a fundamental principle of the stock market. In a Keynesian beauty contest, contestants cannot lose as long as they pick the most popular face. In the stock market, it is very possible to lose even if you pick the most popular stock, in the event of a bankruptcy for instance.
That explains the recent LinkedIn IPO prices, people trying to guess how much the 'average' people value LinkedIn rather than how much you value it (or the value based on certain indicators)
A key result of this analysis is that there's an element of feedback. If you're the only investor in the world who's using this strategy, it may work. But the larger the proportion of investors who do this, the more locally unstable the market will become. It's a far cry from the "perfect, self-correcting market", which we can think of as the zeroth order long-term behaviour. But feedback effects are not sufficiently local that they cannot have destructive effects over a multi-year timescale.
> This would have people pricing shares not based on what they think their fundamental value is, but rather on what they think everyone else thinks their value is, or what everybody else would predict the average assessment of value is.
This is true for people who are buying as speculative investors and focused on short term movements in stock price.
However, that's not everyone by a longshot. There are plenty of people who buy for underlying asset value and dividends.
Warren Buffet has said that The Intelligent Investor is the best book on investing ever written.
A fundamental tenet of value investing is that you're buying a small part of a business, not something separate from the business. Keynes is talking about buying stocks to speculative on short term price movements, which is risky and probably a bad strategy for the vast majority of people. But buying a share of a business that's fundamentally solid and priced attractively is rational, even if the business isn't currently popular.
It seems to me that for companies that pay no dividends (and assuming you would have not nearly enough stock to have a vote that matters) there is no value at all in the stock outside of the stock price. The only relevant metric is what other people are willing to pay to buy your stock. Through some mechanism that I can't undertand, the perception of the stock value is still tied to the performance of the company, so it seems like even people going long are still dealing with their predictions of other peoples valuations.
> for companies that pay no dividends ... there is no value at all in the stock outside of the stock price [but] the perception of the stock value is still tied to the performance of the company
Owning a stock is like owning a fraction of a company. The value of stock comes from the following:
(1) dividends (that is, a fraction of current profits)
(2) assets (that is, a fraction of stuff the company has that could be sold; one of Warren Buffett's strategies is to buy companies that have more assets than their total stock price, such that merely liquidating the company would turn a profit.)
(3) estimated future dividends (a company may not presently be paying dividends, but may be on a clear trajectory to profitability and therefore future dividends)
(4) estimated future assets ("growth" type stocks are in companies that are investing profits in assets rather than paying out dividends. This has certain tax advantages over dividend-paying stocks.)
(5) estimated future valuations (that is, guesses as to what others might pay for the stock later)
Notice that only one of these involves perceptions of others' valuations. The rest are a matter of company performance.
For assets (or future assets), you need to discount them by a probability that the investor could ever actually see that money, either via dividends, or via a 'liquidity event' of one sort or another, since an asset isn't really your asset if you can never touch it. Warren Buffet sees to that by buying controlling stakes in companies, so not only do the companies have assets that could theoretically be liquidated for a profit, but he could actually liquidate the company if he wanted to (or at least, could force it to start selling off assets and returning their value as dividends).
Centuries of financial crises and speculative contagions seem to be enough evidence for me.
If majority of market participants were value investors you simply would not have bubbles to the degree that we have seen historically.
Value investors buy assets at a discount to their intrinsic value and they derive that intrinsic value using conservative assumptions on factors such as liquidation value and future cash flows.
If majority of market participants were value investors you would not have had the Dutch Tulip Mania, South Sea Bubble, or more recently the Dot Com Bubble / pre-GFC highs reached in 2006/2007.
If majority of market participants were value investors you simply would not have bubbles to the degree that we have seen historically
I'd like to see that quantified.
If 60% of investors were value investors and 40% were speculators, would we still get bubbles? I'm thinking yes.
In reality, most people are in the middle. Looking for value, but still mortal, fallible and susceptible to being persuaded that the flavour of the month really is the next big thing.
Not really. Bubbles can easily occur in the absence of credit. I'm not sure how many people were borrowing money in the south seas or tulip bubbles, but probably not that many. Even in the first dot-com bubble (how quaint it sounds to say it now) most of the investing was being done by folks who had money rather than folks who were borrowing it.
The recent housing bubble, on the other hand, yes.
The most common explanation given to economic students is theoretical and relevant to basic rationality:
* investors who are speculative have no interest to become long-term focused, while
* a long-term focused investor who, say, owns a great but over-priced stock has interest in selling it expecting to buy it back afterwards, and then bring it to its long-term value;
* similarly, a long-term investor who sees two company would rationally prefer a average but underpriced stock over a great, but already high one if he expects a higher value.
Because short term always precedes long, considering speculation is rational for either type of investor.
You can check chapter 10 and 11 of Keynes' General Theory… if you want a very detailed personal account of those.(Keynes was the best speculator of his time, and make Cambridge King's College immensely rich doing so.)
I think buying for the long term is a very good thing, but I tend to assume that there aren't many undervalued stocks laying around. Bubbles happen even if there are smart people around who know not to take part, but if there are smart people around its hard for a stock to get too undervalued and I don't care to assume I can do a better job of computing a stocks overall value than professionals do. Sometimes people like Warren Buffet can find gems in the rough, but a large part of that is that they have the resources to swoop in and buy enough stock to sack managers who aren't doing a good job. I can't do that.
I tend to ask myself instead "What do I know that the market doesn't?" I have an engineering education and experience and I can make judgments about a companies products and the long term future of their products that Wall Street investors might not be able to. This is all no guarantee and I keep most of my money in index funds, but its worked out well for me so far.
I don't think the insights offered by TII necessarily disagree with the OP. In fact, I'd say the whole concept of Mr Market chimes in quite well. Both imply a policy of wariness when approaching company values.
My mother taught me that basic strategy for betting on a "horse race" carnival game with the similar design (but the winner was chosen randomply via multiple rolls of a die that represented each "step" along a racetrack)
She taught me to watch the bettors and bet on whichever horse had the least number of shares purchased. If the bets were semi-secret, and we had to engage people ("analysts"?) in conversation about their supposed bets, it would be even more like the real equities marker.
Where are the people that estimate how much of the population classify into level 1, 2, 3 ... and pick choice of the people in the level with the most people.
