I remember reading about it back in college but it seemed very technical. How can someone like me invest using a smart strategy without it being overly complicated?
Hi: The problem with MPT is that it assumed a normal distribution of returns. Without getting too technical- It did not provide for fat tail events: highly improbable events-that have been happening with increasing frequency. That does not mean that MPT is dead, but in a market meltdown all assets, except cash and US Treasuries are highly correlated.
Hi: MPT was developed by Harry Markowitz- who assumed that all investors were rational and risk adverse. Therefore the job of an invesment mgr was to maximize return per unit of risk. This was done through Diversification: having different asset classes, not highly correlated. When some zigged, others would zag. In other words- having bonds, stocks and cash in a portfolio- diversification and non correlation of assets would reduce risk, which was defined as volatility or standard deviation from the mean return.
Would it not be nice if markets followed the normal distribution (that familiar Bell Shaped Curve?) We live in a world of spin doctors. We use the term "revenue enhancements" instead of "taxes." Corporations refer to failed business events as "nonperforming assets." The military refers to civil deaths as "collateral damage," bombs as "vertically deployed antipersonnel devices," and killing the enemy as "servicing the target." Is it any wonder then that economists invented the term "Modern Portfolio Theory" to describe what they would say was the "Rational Market?'
Mark Rubinstein and a US Berkeley colleague calculated the likelihood of a market drop like that of October 19, 1987 in a world of normally distributed price changes as being in the neighborhood of 10 followed by 160 zeros. In other words, that crash should not have happened once every several billion years! It should have been apparent then that the normal distribution does not fit market returns. It still took another 20 years for many academics to acknowledge that reality that the market returns are not normally distributed. Then in 2008-9 the academics and professionals began to talk of "fat tail distributions. I expect that those will be proven just as wrong.
You were better off forgetting Modern Portfolio Theory. It does not work. Read the book "The Myth of the Rational Market" by Justin Fox.
Let me see if I can make this even easier to understand. Modern Portfolio Theory (MPT) is like a see-saw. On one side is risk and the other is reward. The idea is that there is an optimal point where the risk and reward are balanced in a sense. That spot is what they call the efficient frontier. Once you are balanced you have achieved the most "efficient" return for the risk you've taken. Hope that makes sense.
Modern Portfolio Theory basically said that there was a mathematical way to find the efficient frontier ... or the optimal balances between risk and reward, based on some assumptions about the way different asset classes (like stock, bonds, and real assets) behaved. In other words, for each amount of risk you were willing to take, it would be possible to calculate the best mix of asset classes to give the highest expected return. Cool theory, but the devil was in the details... the underlying assumptions, which proved to not be true!)
Here are some of the key assumptions: 1. Fluctuations in asset classes (or the ups and downs of the market prices) followed a Normal Distribution (or the Bell Curve). In other words, small periodic (weekly, monthly, annual) moves occurred frequently, and large moves were less frequent AND followed this bell curve. 2. Asset class behavior in the future would be the same as it was in the past. In other words, if stock prices gained an average of 9.2% per year for the last 20 years, with a certain standard deviation of annual returns, then that's how stock prices would behave in the future. This would allow you to estimate the probability of a certain move in stocks based on past history. 3. Asset class correlations in the future would stay the same as they had in the past. For example, if stock and bond prices moved in opposite directions 60% of the time in the past, that would hold true in the future. 4. Investors are rational, and so everyone will behave to mathematically maximize their long-term returns. As you can see, those assumptions just aren't true in the real world. People panic or get greedy; markets can all move down at once when everyone wants cash, and asset class behavior can change. Stock prices do not actually follow the Normal Distribution, so big moves down actually happen far more often than Modern Portfolio Theory predicted. So, what can be done?
I developed an algorithm for managing my client assets that incorporates the basic ideas of Modern Portfolio Theory, but does not rely on those assumptions to hold true. I then developed a mobile app which uses my algorithms to allocate portfolios based on actual, rather than theoretical market behavior. If you are interested, check it out at www.portfoliowisdom.com and if you like, I'll get your a free copy to evaluate. Dale Beals www.portfoliowisdom.com
Barry are there any alternative theories that work like MPT but properly control for fat tail events?
MPT is not perfect, does not work 100% of the time, but in my opinion diversification is still the best method to reduce risk.