Diversification is a way to reduce risk by investing in a variety of assets or business ventures.
Systematic risk is not diversifiable, while idiosyncratic risk can be reduced or even eliminated.
Portfolio diversification depends on risk-aversion and time horizon, and the portfolio mix must be rebalanced periodically.
Overdiversification/“diworsification” can occur under certain conditions. Business diversification relies on endogenous opportunities, whose value depends on how flexibilities such as timing and expansion options are managed.
To diversify is to do things with variety in order to improve well-being. Diversification is thus a common and fundamental concept in both daily life and business. However, the practice is primarily known as a way of reducing risk by investing in a variety of assets or business ventures. Buying one utility stock in the East coast and one in the West will minimize local shocks, while maintaining roughly the same return as buying either of the two alone. A shop at a resort selling both umbrellas and sunglasses clearly will have a less variable income whether a sunny or a rainy day comes up.
To obtain the optimal strategy of diversification, the risk must be defined and the associated investment opportunities modeled. In addition, the utility or investor’s risk tolerance and investment horizon must be specified. In terms of asset allocation and portfolio choice, the risk is usually defined as the standard deviation of the portfolio return. This measures the variability of the return relative to the expected value of the return. Given a fixed level of expected return, the strategy that generates the minimum variance is preferred. To achieve this, the optimal diversification among the assets will usually be required. The risk tolerance of an investor determines the trade-off between return and risk, as well as the level of risk to take.
Modern Portfolio Theory
Without a formal framework, naive diversification calls for an allocation of an equal amount of money across N assets, and thus it is also known as the 1/N rule. This rule goes back to as early as the fourth century, when Rabbi Issac bar Aha suggested: “One should always divide his wealth into three parts: a third in land, a third in merchandise, and a third ready to hand.” Naive diversification is clearly not optimal in general. For example, when investing in a money market and a stock index, few investors will allocate 50% to the money market.
In 1951 Markowitz published his famous portfolio theory, which provides the optimal portfolio weights on a given N risky assets (stocks) once the expected returns, covariances, and variances of the assets are given, along with the investor’s risk tolerance, in a quadratic utility function. The resulting optimal portfolio is a full diversification with money invested in all of the risky assets. The benefits of diversification depend more on how the assets perform relative to one another than on the number of assets you want to invest. The more the assets do not behave alike—that is, the lower the correlations among them—the more the risk can be minimized by holding the right mix of them.
The optimal portfolio is not risk-free. It is simply the one that has the minimum risk among all possible portfolios of the assets for a given a level of expected return. For any asset, one can decompose its total risk into two components, systematic/market-wide risk and idiosyncratic risk. The optimal portfolio has only market risk, because idiosyncratic risk is diversified away. As a result, there is no point in taking any idiosyncratic risk. But market risk is unavoidable. Intuitively, the return on a suitable portfolio of all stocks in the market has only the market risk, and will not be affected by bad news from some companies, which is likely be offset by good news from others. However, a war, a national disaster, or a global crisis will likely affect the entire portfolio in one direction.
With leverage, the optimal portfolio can theoretically be designed to obtain any desired level of expected return by taking certain necessary risk. The greater the desired expected return on the optimal portfolio, the higher is the risk. Without borrowing and short selling, the diversified portfolio must have an expected return between the highest and the lowest of the asset expected returns. However, the risk is often much smaller than the lowest risk of all the assets.
An efficient portfolio is one that offers either the highest expected return for a given level of risk or the lowest level of risk for a given expected return. The efficient frontier represents that set of portfolios that has the maximum expected return for every given level of risk. No portfolio on the efficient frontier is any better than another. Depending on the investor’s risk tolerance, the investor chooses theoretically one, and only one, efficient portfolio on the frontier.
The investment opportunity set is static in the mean–variance framework underlying the Markowitz portfolio theory. As investment opportunities change over time, many argue for time diversification—that the risk of stocks diminishes with the length of the investment horizon. While this is debatable, the benefit of diversification across assets, and much of the mean–variance theory, carry through into dynamic portfolio choice models with changing investment opportunities. However, due to incomplete information (such as parameter and model uncertainties), trading costs (such as learning and transaction costs), labor income, and solvency conditions, it can be optimal theoretically to underdiversify—to not invest in all assets. Diversification purely for the sake of diversification can cause unnecessary diversification or overdiversification, to end up diworsification i.e. worsening off from bad diversification.
A Stock Investment
Consider an investment in General Motors, or IBM, or the diversified S&P500 Index for 50 years from 1957 to 2007. Examined at a monthly frequency, and based on all the 50 years of data, the estimated expected return on the three assets are 1.12%, 0.84%, and 0.69% per month, and the estimated monthly standard deviations are 7.01%, 7.59%, and 4.13%, respectively. IBM has the highest expected return, with return per unit of risk of 0.16. Although the market has the estimated lowest expected return, its return per unit of risk is higher, 0.17. On the risk-adjusted basis, the market is the best of the three.
In practice, good firms like IBM are not easy to identify ex ante. If one randomly picks a single stock, the average expected return is almost the market return but with much higher risk. The same is true if one randomly chooses a small group of stocks. In fact, back in 1957, IBM, GM, and Eastman Kodak were all blue chip stocks in the famous Dow Jones Index. Eastman Kodak has long gone from the index, and GM it seems is on the verge of being the next to go, with its value drops of more than 75% up to October, 2008. In addition, many firms have gone bankrupt, merged, or been bought over the years. Hence it is important to hold a diversified portfolio and to manage it over time.
Business Diversification: Real Options
With competition, the margin of any business diminishes over time. It is therefore vital for a company to make constant innovations and take on good growth opportunities. All of these activities are closely related to diversification. To enhance existing businesses, a company can diversify geographically in its production and R&D, and diversify vertically to take on more of the functions of the businesses previously run by others. While this increases efficiency and reliability, it also increases the risk exposure of the existing business. A company can, however, diversify horizontally by making new products and opening new markets.
Business diversification is, however, much more complex than stock investment diversification. First, the diversification possibilities are not obvious and have to be studied and developed with resources. Second, the risk and return on a new business are endogenously determined by how it is managed. Third, the benefits of diversification may not show up at the start. This is because existing businesses can be weakened when both management and financial resources are switched to diversification. Also, the new business will typically experience higher risk since the firm’s management has less experience in running it.
Business diversification is almost always sequential. For any project, there are usually many embedded options, such as when to start, whether to expand, and how to switch. Optimal exercise of the options can enhance the value significantly, and so they should be analyzed carefully. Various risk management practices in a company can also be viewed as diversification whereby, to reduce risk, investments are made in financial assets or derivatives to offset the occasional negative payoffs of businesses. However, business diversification can be counterproductive if funds are inefficiently allocated across divisions, if division managers are self-interested, or if the conglomerates, created through mergers of already inefficient firms perhaps, remain inefficient. In addition, diversification typically provides consistent performance with less upside surprises. Academic research finds a diversification discount—that a diversified firm usually trades at a discount relative to a comparable matched portfolio of single-segment firms.
Making It Happen
For asset investments, while money managers can apply modern portfolio theory to diversify the risk with a desired level of return, individual investors can make use of the theory indirectly by investing in portfolios managed by the money managers. In practice this can be done via mutual funds and exchange traded funds. Large and wealthy investors can diversify even more with alternative asset classes such as hedge funds, collectibles (like art works), and exotic investment vehicles sold by large banks. They can diversify over various investment styles, managers, and brokerage accounts. For businesses, diversification means putting managerial and financial resources from your primary business into other opportunities. A small investment in strategic planning and diversification can pay off handsomely later. Vertical diversification may be the first to be started, though horizontal vehicles can be pursued at the same time. Diversification can increase the risk of the existing business but reduce the total risk exposure of the firm. However, the various flexibility options in diversification, such as timing and switching, must be valued carefully and managed efficiently to obtain the maximum diversification benefit.
For asset investments, diversification is an effective tool in reducing the risk of investments in stocks, bonds, and other securities. Utilizing the correlation structure among the assets, idiosyncratic risk can be reduced or even eliminated. For businesses, diversification is a strategic decision. It is vital for a firm’s long-term value creation to identify and manage growth opportunities. Diversification is an important way to manage these opportunities well, reducing risk and ensuring success.