Alpha has become a virtual obsession for investors and the financial commentariat. They seek it, search for it, hunt it and go on a quest for it. Alpha is supposed to be the holy grail of investment. Those who achieve it are alpha masters and, if it were not so politically incorrect, there would no doubt be a rash of self-proclaimed alpha males and possibly females too. There is evidently even a place called Alphaville.

The problem is that alpha does not exist. Or at least not in the way that it is generally discussed.

Alpha is often taken as a synonym for fund manager skill. The temptation to use it in that way is understandable but it should be shunned.

The key point is that alpha is simply a residual. In other words it is the return left over when market return – beta – is accounted for.

There are other factors besides skill that can be included in alpha. For example, it could simply be that beta is not properly measured in calculations. If beta is underestimated, then, by definition, alpha will be overestimated.

A more prosaic alternative is that some alpha is the result of luck. Mathematically this would count as alpha, even though it has nothing to do with skill.

Take a simple analogy. Say that 100 people were asked to toss a coin 10 times with the goal of achieving heads. The average return in this case – assuming the coin is fairly weighted – would be 50%. In other words, the average person would be successful about half of the time. But even if the coin tosses were completely random it is likely that some lucky people would do substantially better than the average. Those individuals could claim to be generating alpha and, strictly speaking, they would be right. But it would clearly have nothing to do with skill. 

This example is adapted from Burton Malkiel’s classic text, A Random Walk Down Wall Street, first published in 1973. The argument here is not that skill in asset management does not exist. It is, rather, that it is misleading to regard alpha as an accurate measure.

In the real world it is certainly difficult to distinguish between skill and luck. It is tricky, if not impossible, to predict with certainly which managers will perform well. 

But even after the event it is not possible to say for sure that a manager that has performed well over a prolonged time period has done so as a result of skill. Even in a completely random world, 10% of funds would be top decile. Some people win lotteries even though the odds may be millions to one against.

It would be best to throw away the comfort blanket that concepts such as alpha provide. The phoney precision only obscures the inherent uncertainty involved. It is possible to search for skilled fund managers but seeking alpha makes no sense.

Special Report: Active Management 

Outsourcing: Avoid the bear traps