11 July 2014
Research suggests that when investors influence each other, the stock market becomes less efficient.
Conventional theory holds that the stock market is efficient and that it does a good job pricing stocks at or close to their fair value, in particular the stocks of large widely followed companies. But could the opposite be true? Could it be that the larger and more followed companies are the less efficiently priced by the market? Could it be that their market value is chronically 20%, 30%, or 40% off from their fair value?
Here is the theory. Assume that there is a finite number of investors, say 1,000 investors, who are active in the stock market and that they each independently derive a value for each stock in the S&P 500. ‘Independently’ here means ‘without sharing thoughts with each other and without letting themselves be influenced by other sources’. Under these admittedly improbable circumstances, the resulting level of the S&P 500 would be quite close to ‘intrinsic value’. We could say that the market would be ‘efficiently’ priced.
Now assume instead that the 1,000 are not working independently but that they influence each other, sharing valuation models, qualitative opinions, price targets, etc. Under these circumstances, the level of the S&P 500 would deviate, in some cases significantly, from its intrinsic value. The market would be inefficiently priced.
This at least is the conclusion you can draw from some recent research on collective decision-making. A recent article titled When Does the Wisdom of the Crowds Turn Into the Madness of the Mob? explains it (my emphasis):
When can we expect a crowd to head us in the right direction, and when can’t we? Recently, researchers have begun to lay out a set of criteria for when to trust the masses.
Democratic decision-making works well when each individual first arrives at his or her conclusion independently. It’s the moment that people start influencing each other beforehand that a crowd can run into trouble.
Philip Ball, writing for BBC Future, describes a 2011 study in which participants were asked to venture educated guesses about a certain quantity, such as the length of the Swiss-Italian border:
“The researchers found that, as the amount of information participants were given about each others guesses increased, the range of their guesses got narrower, and the centre of this range could drift further from the true value. In other words, the groups were tending towards a consensus, to the detriment of accuracy.”
“This finding challenges a common view in management and politics that it is best to seek consensus in group decision making. What you can end up with instead is herding towards a relatively arbitrary position.”
If the research is valid, it debunks the idea that a widely followed stock is efficiently priced. It is not uncommon to hear someone say: “this company is followed by so many people that I have no edge investing in it”. The opposite is almost certainly true: the more widely followed a stock is, and the more ‘influence’ is traded between the participants, the more certain you can be that its market price is wrong, and possibly wrong by a substantial margin.
Take Apple stock for example which is followed by a large number of analysts and investors. When it comes to AAPL price targets, can we say, to paraphrase the article, that the “range of their estimates got narrower, and the center of this range has drifted further from the true value?” And are investors as a group “tending towards a consensus, to the detriment of accuracy?” Investors tend to cluster their price targets not far from the current price which is now $95. But we can theorize that Apple’s intrinsic value is not $100 or $90. It is probably much further from its current market price, say $70 or $120.
Another conclusion can be drawn. When discussing their investment process, fund managers tend to put emphasis on the individual expertise of sector analysts and on their team’s collaborative discussions. In a typical model, the sector analyst will initiate an investment idea and pitch it to a fund manager or to a team who will then reach a decision on how to proceed. In this case, the sector analyst may have been influenced by his peers, by the sell-side and by other sources. And the deciding team members, while searching for a consensus, may have been influenced by each other, by the analyst, and by some willingness to defer to the analyst’s expertise.
If you believe the research described above, this is not the best approach to choosing investments for a portfolio. A better approach would be to have 5 or 10 analysts value the same stock independently, without looking at other sources. It might also be better if these analysts were generalists instead of sector specialists who may be biased in favor of their sector. Once the work is done, there is no point in having any discussions which may prove to be counterproductive. In theory, ‘discussion’ means ‘influence’ and it would result in more bad decisions. It is better to simply look at the valuations derived by these independent analysts. If the average of their price targets is way off the market price, it would be worth initiating a position.