The Portfolio Effect
The Portfolio Effect |
|
The Portfolio Effect Principle: As assets are added to a group (portfolio), the risk of the total portfolio decreases. This will be true as long as the correlation of the asset being added and the portfolio is less than +1.0.
Correlation: The extent to which two items move in concert. Correlation has two components: a direction and magnitude. The direction is shown as a plus or minus sign; the magnitude is shown as a number between zero and one. For example, a correlation of -1.0 indicates that to assets move in opposite directions but by the same percentage amount. That is, when one moves up by 10%, the other moves down by 10%. A correlation of +1.0 means that the two assets move in the same direction and by the same relative amount. A correlation of zero means that the two assets' movements are unrelated or random.
Consider the common stocks of McDonald's Corp. and Wendy's International Corp. Would you expect these two common stocks to typically move in the same direction or in opposite directions? On the one hand, you might say, "Well, they are competitors with one another; therefore if one increases its market share at the expense of the other, I would expect the two to move in opposite directions. So they must have a negative correlation."
On the other hand, there are some more important factors at play. Since the two are in the same industry, they are influenced by the same factors. If the price of beef goes up, the profits of both may go down. If people eat out more often at fast food restaurants, both will benefit and the profits of both will go up. If the minimum wage increases, the costs of both will go up. Overall, due to these characteristics, we would expect that the two stocks would have a positive correlation. But due to the market share issue mentioned above and other factors, the correlation will be less than +1.0.
If we were to compare Intel Corp. to McDonald's, we would expect that the common stocks of the two companies would be somewhat random in their movements relative to one another. After all, they are two entirely different types of companies and are affected by entirely different issues. The factors that affect the revenues and costs of one are very different than the factors that influence the revenues and costs of the other. So we would expect the common stocks to have a correlation close to zero.
Assume that you own McDonald's Corp. stock and would like to reduce the risk, or variability, of your portfolio's returns. If you add a second stock to the portfolio, would it be better to add Wendy's International or Intel?
Assuming that McDonald's and Wendy's have a large positive correlation, Intel would be the better choice. Adding Wendy's to the portfolio would not do much to reduce the variability or risk; the correlation is just too high between the two. When McDonald's stock price decreases a lot, it is likely that Wendy's will decrease a similar amount because the cause of McDonald's decline will likely affect Wendy's in a similar manner. So the risk of the portfolio would be relatively unaffected.
On the other hand, Intel Corp.'s stock would move independently of the McDonald's stock. There will be substantial periods of time when one is going up and the other is going down. This will tend to smooth the fluctuations of the entire portfolio, leading to a lower risk exposure for the investor.
As long as we continue to add stocks that are relatively uncorrelated with the portfolio, the risk of the portfolio will continue to decline. There are limits to this effect however. If we own McDonald's, adding a second stock might cut the variability of the portfolio's price fluctuations almost in half. Adding a third stock likely will have have a lesser effect. Adding a fourth stock will have even a smaller effect. Research studies indicate that almost all of the reduction in risk of a portfolio is reached by the time that the portfolio contains 12 to 15 stocks. Adding additional stocks beyond this point do very little to reduce the risk of the portfolio further.
With the portfolio effect, it is easy to see why corporations like to diversify their product line. Manufacturers of snowmobile engines (a winter product) may decide to manufacture engines for ski boats (a summer product) as well. Manufacturers of swimwear will also produce parkas and sweaters. It is important that the sales of one product be increasing at the time the sales of the other product are decreasing. If the sales of the summer product and the winter product have a -1.0 correlation (i.e., one's sales increase 20% when the other's sales decrease 20%), a graph of the company's total sales would show a straight horizontal line. Obviously, a company would prefer to have stability in its sales as opposed to sales that are much more seasonal.