An enormous debate in asset management centers on the question of whether markets are efficient or not. The answer is important because it determines whether one believes that actively managed investment portfolios can make profits other than by sheer chance. If one believes the market is efficient, then the best strategy is to hold a portfolio that mirrors the market; if not, then the best strategy is to find a smart active manager (or be one!). The fact that some managers have outperformed a simple “buy the index” strategy does not prove anything, unless one can convincingly show that which managers will outperform can be predicted ahead of time. Knowing what numbers worked last on a roulette wheel, for example, won’t help us figure out what numbers come next, if the wheel is fair.
One thing that is sure is that – efficient or not – it is tough to beat a passive index consistently. But what does one mean by “an efficient market?” In general, there appear to be two, possibly three meanings of efficiency. The most common meaning is “informational efficiency,” which essentially means that market prices adjust instantaneously to new information that could inform future prices. Many investors are familiar with in three forms of informational efficiency: weak, semi-strong, and strong forms. The weak form says that any information from past price behavior is already incorporated into the current price and is therefore useless for improving estimates of future returns. The semi-strong form says that all publicly available information (including that extracted from balance sheets and company fundamentals) is reflected in the price. The strong form says that all information, public and private, is reflected in current prices, and has the implication that profitable insider trading is logically impossible.
There is another definition of efficient, which has to do with risk and portfolios. This approach, developed by William Sharpe for CAPM, suggests that the market is efficient because the market – taken as a whole – has no firm-specific risk. It is (by definition), fully diversified, and all firm-specific returns average out to zero. Since firm-specific risk generates no excess return, it is best to hold the market portfolio because it automatically maximizes the anticipated return for the level of risk.
Note that this second concept of market efficiency says nothing about how quickly information makes it into asset or index prices – merely that other combinations of assets necessarily involve a greater degree of asset-specific risk without necessarily getting additional returns.
Finally, there is a view of efficiency that looks at the economic allocation of capital. A capital market is efficient when all available capital moves to its optimal allocation in the economy, which means that capital moves to where it is most productive, and that any changes in the allocation of capital will result in reduced overall productivity. This is a definition of efficiency that is separate from the individual return on assets and may look to economic synergies in capital allocation that may not be obvious simply by looking at total returns. This view of efficiency is separate from the informational efficiency idea, and also distinct from the risk/reward approach.
So when people say that one should hold the market portfolio because the market is efficient, they may be talking about unpredictability, or they may be talking about making sure that the risk will be compensated by higher returns. Investors probably aren’t talking about the efficiency of capital allocation in the economy as a whole, but economists usually are. One key question is whether the two types of investment efficiency are related. In some sense, the informational efficiency argument implies that market timing is difficult, because any information that would allow you to market time is incorporated quickly into prices. The risk efficiency argument implies that improved returns from security selection are difficult, since security selection almost by definition involves taking risks that are theoretically unpaid.