How Allocation Reduces Risk

Recall that an asset class is a broad category of investment, such as stocks, bonds, and real estate.  Each of those classes can be further subdivided into categories.  Stocks, for example, can be divided into geographic categories, such as U.S. Stocks, European Stocks, and emerging market stocks.  Bonds can be divided into government bonds, corporate bonds, taxable bonds, and tax-free bonds.  Real estate can be divided into commercial and residential.

For each of these examples, I’ve only provided a very few possible categories to illustrate the division into categories.  Investors can also divide these categories into styles and sectors.  Styles may include characteristics such as value stocks, growth stocks, and blend stocks.  These are based on the growth characteristics of companies, while other style designations speak to the size of the company.  There are small-cap stocks, mid-cap stocks, and large-cap stocks (cap is short for capitalization, or total market value of all of the company’s stock).  Stocks can be further divided into sectors, which represent the industry in which the company is engaged.  Technology, Energy, and Industrials are examples of stock sectors.  Bonds can be similarly organized by interest rate sensitivity and credit quality.

A well-diversified portfolio may hold several asset classes, and these may be represented by several different categories, sectors, and styles.  The reason we care about all of these divisions and subdivisions is that we need to know how all potential assets in our portfolio will behave in relation to others.  Investment professionals often refer to risk in terms of volatility or variation, but in the world of portfolio management, risk is most often conceived as covariance (correlation).  It is often said that investors should be diversified.  The reason we care about diversification is that a safe portfolio must be composed of uncorrelated assets.  Let’s examine the opposite extreme to underscore the importance of this concept.

If you pay attention to the financial news at all, you no doubt know that technology stocks have been “on fire” for the past few years.  Let’s say you decided to invest in shares of Netflix, Amazon, and NVidia to capture some of these impressive gains.  On March 26, 2018, you would be feeling smug about that decision.  Netflix was up 6.5% for the day, and Amazon was up 4.31%, and NVidia was up 5.08%.  By all metrics, that is an excellent day for any investor.

My point is not for you to go buy a bunch of tech stocks, but to examine how closely related those moves for the day are.  They are not exactly the same, but they are all up, and the magnitude of the change is very similar.  In the terminology of statistics, these price changes are highly correlated.  To see the problem, go back and mentally put negative signs in front of each of those percentage gains.  When we consider the effect of highly correlated losses on our portfolio, the prospect is alarming.

In a very conservative portfolio (given current interest rates), a 6% gain is a good year.  If the above gains were to occur as losses (as they did the trading day before those gains were posted), then the loss could take a year or more to recover from (especially if you do the wrong thing and decide to get out of technology stocks and buy some nice, safe bonds).  If part of your portfolio were invested in long-term treasuries, you would have realized a gain to offset the loss in your technology stocks.

When one asset goes up as another one goes down in value, the assets are said to be negatively correlated.  For offsetting large losses, nothing beats negatively correlated assets.  The bigger the correlation, the more you can depend on this “insurance” working properly.  The problem is that there are no perfect negative correlations in the market.  There are small negative correlations, and usually, the best we can hope for is uncorrelated assets.  When an asset has no correlation with another asset, the assets are said to be independent of each other.

Another important consideration can be the tax liabilities that arise from owning different types of assets.  Taxes can get very complicated, different assets can be classified into different tax brackets, and those brackets can change depending on how long you own the asset.  Since my focus with this book is long-term investing for retirement, I haven’t given taxes much consideration.  Most investors will have their retirement investments in an IRA or 401(k) style account, and you don’t pay taxes on profits in these accounts no matter how you make it.  If you are very wealthy and in a very high tax bracket and are investing outside of a tax-sheltered account, then tax considerations may be paramount.

Many traders out there will tell you “passive” investing strategies are stupid, and that you need to adjust your portfolio to reflect current market conditions.  This is merely market timing by other names, and I hope that I have convinced you of the futility of that endeavor.  If I am your broker and I am paid on commissions, I will encourage you to shift investments quickly because that is how I rack up brokerage fees.  There is no good empirical evidence that any market timing strategy will work.  In 2000, Roger Ibbotson and his colleagues published the definitive study of the relative importance of portfolio allocation versus stock picking in determining asset returns in the Financial Analysts Journal.

The study concluded that more than 90% of the variance in a portfolio’s return is explained by asset allocation, individual asset selection, market timing, and other factors explained the balance.  Note that this bold assertion is measuring (via a coefficient of determination) the relationship between the movement of a portfolio and the movement of the overall market.  In other words, more than 90% of the volatility of a portfolio is due to market movement of the asset classes in which the portfolio is invested.

This is only indirectly related to returns but speaks directly to the most common conceptualization of risk.  As critics have pointed out, the impact of asset allocation on returns depends largely on investing style.  For the active trader, stock picking is the key to explaining variation in returns.  For investors using a long-term passive strategy (only trading during periodic rebalancing according to an investment plan), the asset allocation decision is by far the most important.


[ Back | Contents | Next ]

Last Updated: 6/25/2018