Volume 19: The Efficient Market Hypothesis
It's Why You (Probably) Can't Beat the Market
“The stock market is a device for transferring money from the impatient to the patient,” – Warren Buffett
As a society, we have decided to invest our time, our currency, and some of our greatest minds towards the task of beating the market. Teams of Ph. D’s sit in New York, Chicago, Greenwich and elsewhere, looking for any tiny edge they can exploit. They combine form a giant machine, taking in information, spitting out predictions for asset prices.
As our knowledge base becomes larger - satellite pictures of Iowa cornfields, micro-data on consumer spending, professional “Fed Watchers” – this machine is able to incorporate more information, faster. With all of this effort, we do a pretty good job; stocks, bonds, commodities and foreign exchange rates must cost just about what they are worth. But this, itself, is a problem. If you can only buy or sell something for what it’s worth, then there is no way to profit. If there were no way to profit, then all of this effort would be wasted. Something doesn’t add up.
This is the Efficient Market Hypothesis, and its paradox.
The Efficient Market Hypothesis is wrong. If you can figure out how and when it’s wrong, you’ll soon find yourself a billionaire.
What is the Efficient Market Hypothesis?
Do professional investors beat the market?
How does anybody make any money?
What is the Efficient Market Hypothesis?
Classical economics is the explanation of production, consumption and transfer of wealth. It predicts how much the price and quantity of apples will change after a weak harvest. It gives direction for how changes in tax policy will affect the economy.[1] It tackles the thorny issues involved when governments regulate the crossing of otherwise arbitrary lines, otherwise known as international trade.
As international capital markets became larger and more sophisticated in the early 20th century, we realized that financial assets don’t always follow the same economic rules as physical goods. To study these differences, Finance Theory began to forge its way as a separate discipline. The Efficient Market Hypothesis, or EMH, belongs to Finance Theory.[2]
In 1950, despite formal financial markets having existed for over 350 years, we still had little theoretical understanding of how assets were (or should be) priced.[3] We knew it had something to do with risk; investors demand higher expected returns for a garage-based start-up than for a U.S. Treasury Bond. But we knew that not all risk was worth of reward. Just because the casino game of roulette has massive risk doesn’t mean its return is better than other investments. We also knew that diversification was generally a good idea. As early as 1615, Sancho Panza told us that a wise man does not keep all his eggs in one basket. But that’s just about all we knew.[4]
A large, early step in the development of finance theory was portfolio optimization. Advanced by Harry Markowitz and others in the early 1950s, portfolio optimization uses correlation to increase returns without increasing risk. Holding a portfolio split between two assets generally has less risk than a portfolio holding only one.[5] Adding additional assets allows even better portfolios. Taking into account the entire investment universe, an optimal portfolio can be constructed for any desired amount of risk. Because this is a limit which theoretically can not be exceeded, these portfolios are called the efficient frontier.
Figure 1 - The Efficient Frontier and Capital Allocation Line
The next step towards understanding asset prices was the development of the Capital Asset Pricing Model (CAPM). Figure 1 shows the efficient frontier; however, you’ll see an additional point, representing the return possible from a risk-free portfolio. Investors can put any or all of their funds in this portfolio. A line starting at this point will be tangent to our efficient frontier at a single point, which we call the tangency portfolio. An investment of some of our assets in this tangency portfolio, with the rest in the risk-free asset, will outperform any other investment. Each investor can choose their own risk tolerance. The portfolios so described are on the capital allocation line, or CAL.[6]
But, therein lies the rub. Taken to an illogical extreme, all an investor needs to do is identify their personal tolerance for risk, feed it into an algorithm, and invest in the resulting portfolio. These investments won’t be based on whether a company is well run, or if people actually want its products, but rather market parameters like volatility and correlation.[7] The results of the algorithm will be optimized, in that, for a given level of risk, the portfolio will have an expected return greater than or equal to any other portfolio. This is a serious issue for mutual fund managers, asset allocators, and other who charge millions to advise clients on portfolio construction.
The Capital Asset Pricing Model and Capital Allocation Line are theoretical constructs; there is no perfect algorithm for real-world markets. But, even if they are vast oversimplifications, they still call into question the vast resources we put into security analysis. If these assumptions can eliminate literally every reason for such analysis, then perhaps the real-world value is still far less than often assumed. This, finally, brings us to the EMH.[8]
When we say that markets are efficient, we mean that their prices incorporate all available information (or a specific subset thereof). We can use an example from sports gambling, which is as useful as it was way back in Volume 7. Consider an NFL game: the Patriots are playing the Packers. Let’s say the initial “market” of gamblers assume that each team has a 50% chance of winning.[9] Then, Tom Brady blows out his knee in practice; it is immediately reported to everybody. The odds will move, with the Packers now being heavily favored. The market was efficient with respect to this new information; it was immediately incorporated in the price. Alternatively, let’s say Tom called only me and said that he was going to sit on Sunday.[10] I would be able to bet against the Pats at very favorable odds; the market won’t really move right away, because nobody else knows. In this situation, the market was not efficient.
We can take another example from financial markets proper: a company reporting quarterly earnings. If a company reports good earnings, the stock price will go up before you could buy any shares. The positive information contained within the announcement is therefore incorporated almost immediately, which prevents you from taking advantage of it. Even further, the nothing of “good earnings” will be based on the expectations of the market overall. This means that not only the earnings report, but also the in-depth studies of many professional analysts are also incorporated in the stock price.
These examples demonstrate a major reason why asset prices change: the market receives new information. This is the essence of market efficiency. However, we’ve also seen that markets are not equally efficient with respect to different types of information. For this reason, there are three forms of the EMH and we are, finally, ready to state them. Recall, however, that no version of the EMH actually describes real markets.[11]
EMH Weak Form: Asset prices incorporate all past price and volume data.
If you flip a coin ten times, getting ten heads, it is tempting to think that on the 11th roll, another head is the most likely result. If the coin is fair, this is of course not true. Similarly, some people think that stocks that have been going up will continue to go up and that stocks that have been going down will continue to go down. If the Weak Form of the EMH were true, so-called momentum trading is no better than a coin flip.
Using “charts” of past prices and data to predict future movement is generally called technical analysis. There is a massive amount of literature showing that making money via technical analysis is, really, really difficult. That being said, there are numerous billions of dollars invested in such strategies.[12] For the first time, we’ve found an apparent paradox with in the EMH.
EMH Semi-Strong Form: Asset prices incorporate all publicly available information.
A lot of people buy stocks because they think its underlying company will be successful. There are literally 24-hour television channels, CNBC being the most prominent, theoretically dedicated to informing the public as to which companies are the good ones. Investment banks, mutual fund managers, hedge funds and other investment managers employ armies of equity researchers to understand and predict the future results and stock prices of corporations. If the Semi-Strong Form were true, all of this effort would be in vain.
Those who make investment decisions based on balance sheets, management credibility, or even whether they saw a full parking lot at a given store, are participating in fundamental analysis. There is, again, a lot of research showing that fundamental analysis doesn’t work very well. Specifically, over the long term, few active equity managers outperform broad-based equity indices.[13] This again brings us back to the paradox; if fundamental analysis is pointless, why do people bother doing it?
EMH Strong Form: Asset prices incorporate all available information.
If you are working at an investment bank, and have non-public knowledge that a client is about to be bought by another company for a large premium, you might be tempted to go and buy some of the stock before the news goes public. Do not do this, it is very illegal, and you are very likely to get caught. And if the Strong Form were true, it would be pointless anyway.
Those who trade on material, non-public information are said to be engaging in insider trading. In equity markets, insider trading is illegal because it can be used to earn outsized profits without taking financial risk. This isn’t only unfair, but would also poison faith in the market.[14] In these markets, the Strong Form is definitionally not true; trading on insider information is expected to result in profits.
There are markets, specifically commodities and foreign exchange, that do not prohibit insider trading. They couldn’t operate otherwise; corn growers always have insider knowledge of future corn prices. A prohibition on their use of this knowledge would be a prohibition on producers trading corn, whic