Part 5 (1/2)

This three-year sequence is a pretty typical one. Let's start with 1998. In the first place, a diversified portfolio did reasonably well in that year. U.S. large stocks did the best, but REITs lost a lot of money. Many investors got discouraged that year and sold their REITs. They were soon sorry because by 2000, stock returns were generally poor and REITs were the only stock a.s.set with superlative returns. Foreign and U.S. large stocks, which delivered excellent returns in the first two years, took a nosedive in 2000.

The key is to ignore the year-to-year relative performance of the individual a.s.set cla.s.ses-their behavior usually averages out over the years-it is the long-term long-term behavior of your whole portfolio that matters, not its day-to-day variation. If you cannot help focusing on the performance of the individual a.s.set cla.s.ses in your portfolio, at least do so only over as long a period as possible. behavior of your whole portfolio that matters, not its day-to-day variation. If you cannot help focusing on the performance of the individual a.s.set cla.s.ses in your portfolio, at least do so only over as long a period as possible.

With training and experience, most investors take these normal a.s.set cla.s.s ups and downs in stride. (There is even a way to take advantage of them, which we'll discuss later in the chapter.) But some investors cannot. If you are such an individual and become upset when one of your a.s.set cla.s.ses does poorly, even when the rest of your portfolio is doing well, then you should not be managing your own money. I can guarantee you that each and every year you will have at least one or two poorly performing a.s.sets. And in some years, like 2000, most will behave miserably.

If you cannot handle the routine a.s.set cla.s.s volatility inherent in the capital markets, then you should have a reputable financial advisor making your investment decisions. Your decisions will forever be clouded by your emotional responses to normal market activity.

Our exploration of the a.s.set allocation process will proceed in several steps. We'll start with the most important allocation question of all: the decision of how much of your capital to put at risk.

Step One: Risky a.s.sets, Riskless a.s.sets Distilled to its essence, there are only two kinds of financial a.s.sets: those with high returns and high risks, and those with low returns and low risks. The behavior of your portfolio is determined mainly by your mix of the two. As we learned in Chapter 1 Chapter 1, all stocks are risky a.s.sets, as are long-term bonds. The only truly riskless a.s.sets are short-term, high-quality debt instruments: Treasury bills and notes, high-grade short-term corporate bonds, certificates of deposit (CDs), and short-term munic.i.p.al paper. To be considered riskless, their maturity should be less than five years, so that their value is not unduly affected by inflation and interest rates. Some have recently argued that Treasury Inflation Protected Securities (TIPS) should also be considered riskless, in spite of their long maturities, because they are not negatively affected by inflation.

What we'll be doing for the rest of this chapter is setting up a ”laboratory” in which we create portfolios composed of various kinds of a.s.sets in order to see what happens to them as the market fluctuates. How we compute the behavior of these portfolios is beyond the scope of this book; for those few of you who are interested, I suggest that you read the first five chapters of my earlier book, The Intelligent a.s.set Allocator. The Intelligent a.s.set Allocator. Suffice it to say that it is possible to simulate with great accuracy the Suffice it to say that it is possible to simulate with great accuracy the historical historical behavior of portfolios consisting of many a.s.sets. Keep in mind that this is not the same as predicting the behavior of portfolios consisting of many a.s.sets. Keep in mind that this is not the same as predicting the future future behavior of any a.s.set mix. As we discussed in the first chapter, historical returns are a good predictor of future risk, but not necessarily of future return. behavior of any a.s.set mix. As we discussed in the first chapter, historical returns are a good predictor of future risk, but not necessarily of future return.

Let's start with the simplest portfolios: mixtures of stocks and T-bills. I've plotted the returns of Treasury bills, U.S. stocks, as well as 25/75, 50/50, and 75/25 mixes of the two, in Figures 4-1 Figures 4-1 through through 4-5 4-5. In order to give an accurate idea of the risks of each portfolio, I've shown them on the same scale.

As you can see, when we increase the ratio of stocks, the amount lost in the worst years increases. This is the face of risk. In Table 4-1 Table 4-1, I've tabulated the return, as well as the damage, in the 197374 bear markets for a wide range of bill/stock combinations. Finally, in Figure 4-6 Figure 4-6, I've plotted the long-term returns of each of these portfolios versus their performance in 19731974.

Figure 4-6 provides the conceptual heart of this chapter, and it's worth dwelling on for a few minutes. What you are looking at is a map of portfolio return versus risk. The numbers along the left-hand edge of the vertical axis represent the annualized portfolio returns. The higher up on the page a portfolio lies, the higher its return. The numbers on the horizontal axis, at the bottom of the graph, represent risk. The further off to the left a portfolio lies, the more money it lost in 197374, and the riskier it is likely to be in the future. provides the conceptual heart of this chapter, and it's worth dwelling on for a few minutes. What you are looking at is a map of portfolio return versus risk. The numbers along the left-hand edge of the vertical axis represent the annualized portfolio returns. The higher up on the page a portfolio lies, the higher its return. The numbers on the horizontal axis, at the bottom of the graph, represent risk. The further off to the left a portfolio lies, the more money it lost in 197374, and the riskier it is likely to be in the future.

[image]

Figure 4-1. All Treasury bill annual return, 19012000. ( All Treasury bill annual return, 19012000. (Source: Jeremy Siegel.) Jeremy Siegel.) [image]

Figure 4-2. Mix of 25% stock/75% Treasury bill annual returns, 19012000. ( Mix of 25% stock/75% Treasury bill annual returns, 19012000. (Source: Jeremy Siegel.) Jeremy Siegel.) [image]

Figure 4-3. Mix of 50% stock/50% Treasury bill annual returns, 19012000. ( Mix of 50% stock/50% Treasury bill annual returns, 19012000. (Source: Jeremy Siegel.) Jeremy Siegel.) [image]

Figure 4-4. Mix of 75% Stock/25% Treasury bill annual returns, 19012000. ( Mix of 75% Stock/25% Treasury bill annual returns, 19012000. (Source: Jeremy Siegel.) Jeremy Siegel.) [image]

Figure 4-5. All-stock annual returns, 19012000. ( All-stock annual returns, 19012000. (Source: Jeremy Siegel.) Jeremy Siegel.) It's important to clear up a bit of confusing terminology first. Until this point in the book, we've used two designations for fixed-income securities: bonds and bills, referring to long- and short-duration obligations, respectively. Bonds and bills are also different in one other respect: bonds most often yield regular interest, whereas bills do not-they are simply bought at a discount and redeemed at face value. The most common kinds of bills in everyday use are Treasury bills and commercial paper, the latter issued by corporations.

Long-duration bonds are generally a sucker's bet-they are quite volatile, extremely vulnerable to the ravages of inflation, and have low long-term returns. For this reason, they tend to be bad actors in a portfolio. Most experts recommend keeping your bond maturities short-certainly less than ten years, and preferably less than five. From now on, when we talk about ”stocks and bonds,” what we mean by the latter is any debt security with a maturity of less than five to ten years-T-bills and notes, money market funds, CDs, and short-term corporate, government agency, and munic.i.p.al bonds. For the purposes of this book, when we use the term ”bonds” we are intentionally excluding long-term treasuries and corporate bonds, as these do not have an acceptable return/risk profile. For the purposes of this book, when we use the term ”bonds” we are intentionally excluding long-term treasuries and corporate bonds, as these do not have an acceptable return/risk profile. I'll admit that this is a bit confusing. A more accurate designation would be ”stocks and relatively short-term fixed-income instruments,” but this wording is unwieldy. I'll admit that this is a bit confusing. A more accurate designation would be ”stocks and relatively short-term fixed-income instruments,” but this wording is unwieldy.

Table 4-1. 19012000, 100-Year Annualized Return versus 19731974 Bear Market Return 19012000, 100-Year Annualized Return versus 19731974 Bear Market Return [image]

The data in Table 4-1 Table 4-1 and the plot in and the plot in Figure 4-6 Figure 4-6 vividly portray the tradeoff between risk and return. The key point is this: the choice between stocks and bonds is not an either/or problem. Instead, the vital first step in portfolio strategy is to a.s.sess your risk tolerance. This will, in turn, determine your overall balance between risky and riskless a.s.sets-that is, between stocks and short-term bonds and bills. vividly portray the tradeoff between risk and return. The key point is this: the choice between stocks and bonds is not an either/or problem. Instead, the vital first step in portfolio strategy is to a.s.sess your risk tolerance. This will, in turn, determine your overall balance between risky and riskless a.s.sets-that is, between stocks and short-term bonds and bills.

Many investors start at the opposite end of the problem-by deciding upon the amount of return they require to meet their retirement, educational, life style, or housing goals. This is a mistake. If your portfolio risk exceeds your tolerance for loss, there is a high likelihood that you will abandon your plan when the going gets rough. That is not to say that your return requirements are immaterial. For example, if you have saved a large amount for retirement and do not plan to leave a large estate for your heirs or to charity, you may require a very low return to meet your ongoing financial needs. In that case, there would be little sense in choosing a high risk/return mix, no matter how great your risk tolerance.

[image]

Figure 4-6. Portfolio risk versus return of bill/stock mixes, 19012000. Portfolio risk versus return of bill/stock mixes, 19012000.

There's another factor to consider here as well, and that's the probability that stock returns may be lower in the future than they have been in the past. The slope of the portfolio curve in Figure 4-6 Figure 4-6 is steep-in other words, in the twentieth century, there was a generous reward for bearing additional portfolio risk. It is possible, for example, that the future risk/reward plot may look something like is steep-in other words, in the twentieth century, there was a generous reward for bearing additional portfolio risk. It is possible, for example, that the future risk/reward plot may look something like Figure 4-7 Figure 4-7, with a much lower difference in returns between risky and risk-free investments. In this ill.u.s.tration, I've a.s.sumed a 7% return for stocks and a 5.5% return for bonds. In such a world, it makes little sense to take the high risk of an all-stock portfolio.

Finally, it cannot be stressed enough that between planning and execution lies a yawning chasm. It is one thing to coolly design a portfolio strategy on a sheet of paper or computer monitor, and quite another to actually deploy it. Thinking about the possibility of losing 30% of your capital is like training for an aircraft crash-landing in a simulator; the real thing is a good deal more unpleasant. If you are just starting out on your investment journey, err on the side of conservatism. It is much better to underestimate your risk tolerance at an early age and adjust your risk exposure upwards later than to bite off more than you can chew up front.

[image]

Figure 4-7. Likely future portfolio risks/returns. Likely future portfolio risks/returns.

Millions of investors in the 1920s and 1960s thought that they could tolerate a high exposure to stocks. In both cases, the crashes that followed drove most of them from the equity markets for almost a generation. Since the risk of your portfolio is directly related to the percentage of stocks held, it is better that you begin your investment career with a relatively small percentage of stocks. This flies directly in the face of one of the prime tenets of financial planning conventional wisdom: that young investors should invest aggressively, since they have decades to make up their losses. The problem with an early aggressive strategy is that you cannot make up your losses if you permanently flee the stock market because of them.

This all adds up to one of the central points of a.s.set allocation: Unless you are absolutely certain of your risk tolerance, you should probably err on the low side in your exposure to stocks.

Step Two: Defining the Global Stock Mix Why diversify abroad? Because foreign stocks often zig when domestic markets zag, or at least may not zig as much. Let's look at the most recent data.

In the early part of this century, the international capital markets were a good deal more integrated than they are now. It was commonplace for an Englishman to buy American bonds or French stocks, and there were few barriers to cross-border capital flow. The two World Wars changed that; the international flow of capital recovered only slowly afterwards. The modern history of international diversification properly begins in 1969, with the inception of Morgan Stanley's EAFE (Europe, Australasia, and Far East) Index. As of year-end 2000, there is a 32-year track record of accurate foreign returns. For the period, this index shows an 11.89% annualized return for foreign investing, versus 12.17% for the S&P 500.

Why invest in foreign stocks if their returns are the same, or perhaps even less than U.S. stocks? There are two reasons: risk and return. In Figure 4-8 Figure 4-8, I've plotted the annual returns of the two indexes. Note how there can be a considerable difference in return between the two in any given year. Particularly note that during 1973 and 1974, the EAFE lost less than the S&P: a total 33.16% loss for the EAFE versus a 37.24% loss for the S&P 500. What this means is that foreign investing provided a bit of cus.h.i.+on to the global investor.

[image]

Figure 4-8. Returns for S&P 500 and foreign stocks, 19622000. Returns for S&P 500 and foreign stocks, 19622000.

An even more vivid case for diversifying into foreign stocks is made by looking at returns decade by decade, as shown in Figure 4-9 Figure 4-9. Notice how during the 1970s, the return of the S&P 500 was less than inflation-that is, it had a negative real return-whereas the EAFE beat inflation handily. You'll also see that the EAFE beat the S&P 500 by a similar margin in the 1980s.

Thus, for a full two decades you would have been very happy with global diversification. This would have been particularly true if these two decades had been your retirement years, since a U.S.-only portfolio would have very likely run out of money due to its relatively low returns. In the 1990s, the law of averages finally caught up with foreign stocks, souring many on global diversification.

Despite the slightly lower rewards of foreign stocks, the most powerful argument, paradoxically enough, can actually be made on the basis of return. Most investors do not simply select an initial allocation and let it run for decades without adjustment. Because of the varying returns of different a.s.sets over the years, portfolios must be ”rebalanced.” To see what rebalancing means, let's look at the two-year period from 1985 to 1986.

[image]

Figure 4-9. S&P 500, EAFE, and inflation, by decade. ( S&P 500, EAFE, and inflation, by decade. (Source: Morningstar Inc.) Morningstar Inc.) The overall return of the S&P 500 for those years was quite high-57%-but the return of the EAFE was off the charts-166%! Had you started with a 50/50 portfolio at the beginning of the period, at the end, it would have been 63% foreign and 37% domestic. Rebalancing the portfolio means selling enough of the better performing a.s.set (in this case, the EAFE) and with the proceeds buying the worse performing a.s.set (the S&P 500) to bring the allocation back to the 50/50 policy.

Had you rebalanced a 50/50 S&P 500/EAFE portfolio every two years between 1969 and 2000, it would have returned 12.62%. This was almost one-half percent better than the best-performing a.s.set, the S&P 500. Why? Because when you rebalance back to your policy allocation (your original 50/50 plan), you are generally selling high (the best performer) and buying low (the worst performer). So, over the long haul, international diversification not only reduces risk, but it may also increase return. But be warned: as the past decade has clearly taught us, foreign diversification is not a free lunch, especially if your time horizon is less than 15 or 20 years.

Until recently, the average U.S. investor did not have to worry about diversifying abroad-it simply wasn't an option. Although domestic investors have been able to purchase foreign stocks for more than a century, in practice this was expensive, c.u.mbersome, and awkward; it could only be done one stock at a time. Although the first U.S.-based international fund opened its doors almost five decades ago, it wasn't until the early 1980s that these vehicles became widely available. In 1990, the Vanguard Group made available the first easily accessible, low-cost indexed foreign funds.

What is the proper allocation to foreign stocks? Here we run into an enormous problem-one that makes even the most devout believer in efficient markets a bit queasy. The rub is that the total market cap of non-U.S. stocks is about $20 trillion versus only $13 trillion for the U.S. market. If you believe that the global market is efficient, then you should own every stock in the world in cap-weighted fas.h.i.+on, meaning that foreign companies would comprise 60% of your stock exposure. This is more than even the most enthusiastic proponents of international diversification can swallow.

So what's a reasonable foreign allocation? Certainly less than 50% of your stock pool. For starters, foreign stocks are more volatile, in general, than domestic stocks on a year-by-year basis. Second, they are more expensive to own and trade. For example, the Vanguard Group's foreign index funds, on average, incur about 0.20% more in annual expenses than their domestic index funds. Finally, a small portion of the dividends of foreign stocks are taxed by their national governments. Although these taxes are deductible on your tax returns, this deduction does not apply to retirement accounts. Here, it is lost money.

Experts differ on the ”optimal” foreign stock exposure, but most agree it should be greater than 15% of your stock holdings and less than 40%. Exactly how much foreign exposure you can tolerate hinges on how much ”tracking error” (the difference between the performance of your portfolio and the S&P 500) you can bear. Take a look again at Figure 4-9 Figure 4-9. An investor with a high foreign exposure would have suffered accordingly in the nineties. Although their returns would have been satisfying, they would have been much less than those obtained by their neighbors who had not diversified. So although the long-term return of a globally diversified stock portfolio should be slightly higher than a purely domestic one, there will be periods lasting as long as 10 or 15 years when the global portfolio will do worse.

If this temporary shortfall relative to the S&P 500-tracking error-bothers you greatly, then perhaps you should keep your foreign exposure relatively low. If it does not bother you at all, then you may be able to stomach as much as 40% in foreign stocks. But whatever allocation you settle on, the key is to stick with it stick with it through thick and thin, including rebalancing back to your target percentage on a regular basis. through thick and thin, including rebalancing back to your target percentage on a regular basis.

Step Three: Size and Value Steps one and two-the stock/bond and domestic/foreign decisions-const.i.tute a.s.set allocation's heavy lifting. Once you've answered them, you're 80% of the way home. If you're lazy or just plain not interested, you can actually get by with only three a.s.set cla.s.ses, and thus, three mutual funds: the total U.S. stock market, foreign stocks, and short-term bonds. That's it-done.

However, there are a few relatively simple extra portfolio wrinkles that are worth incorporating into your a.s.set allocation repertoire. We've already talked about the extra return offered by value stocks and small stocks. The diversification benefits of small stocks and value stocks are less certain. For example, during the 197374 bear market, value stocks did much better than growth stocks; the former lost only 23% versus 37% for the latter. But during the 192932 bear market, value stocks lost 78% of their worth, versus ”only” 64% for growth stocks. The academicians who have most closely examined the value effect-Fama and French-insist that the higher return of value stocks reflects the fact that these companies are riskier than growth stocks because they are weaker and thus more vulnerable in hard times. Fama and French's theory is consistent with stock performance during the 192932 bear market.