In Asset Allocation: Part 1, I discussed some of the thinking behind diversifying your portfolio among stocks, bonds and cash. You can accomplish this quite simply with 2 or 3 low cost index mutual funds, and some highly respected financial authors and researchers recommend keeping it simple, and doing exactly this. However, other highly respected financial advisors/authors/researchers recommend splitting your portfolio into more asset classes, and cite research that demonstrates that this has often (but not always) provided higher return with lower risk over the long run. This is what I'll be discussing in this post.
Modern Portfolio Theory
Starting in the early 1950s, academic researchers started discovering that you could combine several different risky assets into a portfolio while reducing the overall risk of the portfolio. The term you might hear to describe this is Modern Portfolio Theory (MPT). To read an in depth article on MPT, see the wikipedia article on MPT.
Asset Class Correlation
Without getting into the theory, let's look at a simple (but unrealistic) example of how combining risky assets can lower the risk of a portfolio.
Assume your portfolio initially consisted of $100 worth of asset A, that alternately increases by 10% one year, then decreases by 5% the next. Over the years, the value of this portfolio would increase, but it would be a bumpy ride (see the blue line in chart below). Now assume you have access to asset B, which performs exactly like asset A except that it goes down when asset A goes up, and it goes up when asset A goes down (red line in chart below). If you split your investment evenly between assets A and B (starting with $50 each), the ups and downs of these 2 assets offset each other, smoothing out the overall performance of your portfolio (orange line in chart below). Note the smooth uptrend of the A+B portfolio compared to the ups and downs of either asset A or B alone.
Correlation ranges from -1 for perfectly negatively correlated assets (as in the above example) to +1 for perfectly correlated assets (assets that move up and down exactly the same). In reality, one attempts to construct a portfolio of asset classes with correlations less than 1 (the lower the better), but that all tend to increase in value over long periods of time. For these less than perfectly correleated assets, sometimes asset A may go down when asset B goes up (as with negative correlation), and at other times they may both move in the same direction, but by different amounts (positive correlation, but less than 1). Modern Portfolio Theory suggests that you should attempt to add an asset class to your portfolio if it has low correlation to the other asset classes in your portfolio. The lower the correlation, the better the potential for increasing returns while lowering volatility.
Note that volatility (how often and how much the value of an asset changes) is commonly used as a measure of risk in investing. Although one can argue that this isn't the best measure of risk, it's certainly an indicator of how risky an investment feels in the short term. Most people would prefer a steady increase in the value of their portfolio rather than a portfolio that increased by the same amount, but that went up and down a lot along the way.
Even in the simple asset allocation described in Part 1, we are taking advantage of the low correlation between stocks and bonds to reduce the volatility of our portfolio. Of course, since bonds tend to be a less risky (i.e., less volatile) asset, we're also reducing the volatility simply by adding a less risky asset to the portfolio.
However, here we're going to explore combining asset classes that are as risky or even riskier than an all U.S. stock fund to potentially increase our long term return, while lowering our risk, or at least without increasing it very much. First, what are the different asset classes we might consider, other than US stocks, bonds and cash?
The first additional asset class to consider is non-US stocks, often referred to as foreign or international stocks (I usually refer to them as international). Although non-US and US stocks often perform similarly, they sometimes perform quite differently. There have been many periods where adding some international stocks to your portfolio would've provided a higher return while smoothing out the volatility (reducing the risk). Of course there have been some fairly long periods of time when US stocks outperformed international stocks, and adding international stocks would've reduced your return (sometimes also reducing volatility, sometimes not).
Most advisors I respect recommend holding at least 20% in international stocks (i.e., a low cost international stock fund), and some recommend holding as much as 50% of your portfolio in these stocks. I may go into the arguments for holding more or less international in another post, but I tend to be more convinced by the arguments to hold more. I'd recommend a minimum of 30% in international, and would consider as much as 50%.
The next thing to consider is that non-US stocks can be divided into developed and emerging markets, each of which can be considered a separate asset class. Developed markets are those in stable countries with well developed markets; e.g., Japan, U.K., Germany, France, Australia, etc. Emerging markets are those with less developed markets, but that are moving in the direction of becoming developed markets. Examples of emerging markets are China, India, Brazil, and Russia.
Emerging markets make up about 24% of the total international market as of 12/31/2009. You will automatically own this 24% of emerging markets if you buy the Vanguard Total International Stock Index Fund or the Vanguard FTSE All-World Ex-US fund. However, you'll own only the developed markets if you own the Fidelity Spartan International Index Fund or the Schwab International Index Fund. So, if you want to own emerging markets (which I recommend), you'll need to be sure to investigate the portfolio of the international fund you own or are considering purchasing. If it doesn't include emerging markets, you will need to invest in a separate emerging markets fund to get exposure to this asset class.
Emerging markets are generally considered riskier than developed markets, since they are in countries that are less stable politically and economically. This contributes to more volatility in the prices of emerging market stocks and stock funds. Most evidence I've seen indicates that long term investors have been rewarded for adding emerging markets to their portfolios.
Value and Small Cap
Another way that stocks are commonly split into different asset classes is by considering size and valuation. Size is measured by the total value of a company's stock (referred to as capitalization or just "cap" for short). Generally stocks are classified as large cap, mid cap or small cap. In terms of valuation, stocks are broken into growth, core (or blend) and value. Growth stocks tend to be growing quickly, and cost more, while value stocks tend to be growing less quickly (or not at all), and cost less.
Some authors portray growth stocks as good companies but bad investments, and value stocks as bad companies but good investments. The thinking goes that investors have lower expectations for value stocks (bad companies), and that the surprises tend to be on the upside; e.g., a company turns around its poor performance and starts performing well, thus raising the stock price. Conversely, high expectations for growth stocks (good companies) tend to lead to high stock prices, and the surprises tend to be on the downside; e.g., a company does not grow as fast as expected, and the stock price falls.
Research has shown that small cap stocks have outperformed large cap stocks, and value stocks have outperformed growth stocks over the last 80 years, and often over various 10-20 year periods. However, there have been long periods (10-20 years) where the opposite has been true (i.e., growth outperformed value or large outperformed small).
Many highly respected financial advisors/authors (e.g., Larry Swedroe, William Bernstein, Rick Ferri) believe that the research showing the outperformance of small cap and value stocks is convincing, and that your portfolio should include more of these stocks than a Total U.S. Stock market fund includes. On the other hand, other highly respected financial authors (e.g., John Bogle, Burton Malkiel, Charles Ellis) believe that you'll do just fine sticking to total market funds (whether US or international).
As with all investments, part of the reason small cap and value stocks tend to have higher long term returns is that they are riskier. Small companies (small cap) and companies that are struggling (value) are more likely to go out of business than large companies (large cap) and companies that are growing quickly (growth). In the long run, investors have generally (but not always) been rewarded with higher returns in these asset classes, while enduring more volatility along the way.
For the 10 years ending December 2008, here are the approximate total returns of several asset classes discussed above (as shown on page 15 of The Investor's Manifesto by William Bernstein):
- U.S. Large Blend: -14%
- U.S. Large Value: + 24%
- U.S. Small Value: +103%
- International Large Blend: +13%
- Emerging Markets Large Blend: +148%
I tend to come down more on the side of adding the additional asset classes, but only if you are willing to deal with the additional complexity of owning more funds, and if you have the discipline and will to rebalance methodically (i.e., buy more of the funds that have gone down, and sell some of the funds that have gone up). You also have to be willing to live with the higher volatility of these other asset classes, noting that the overall volatility of your portfolio will usually be much less than the volatility of these individual asset classes.
The dimensions of Small/Large and Growth/Value are often shown in a 9 square style grid, with Large Growth stocks in the top right box, and Small Value stocks in the bottom left box. If you have a Vanguard account, Vanguard's Portfolio Analysis tool can show how the diversified U.S. stock funds in your portfolio are distributed among the 9 boxes in the style grid.
Large Value Large Blend Large Growth
Mid Value Mid Blend Mid Growth
Small Value Small Blend Small Growth
Another asset class that is often recommended is Real Estate Investment Trusts (REITs -- pronounced "reets"). These are companies that own apartment buildings, shopping centers, and other income generating real estate, and that trade like stocks. As with the other asset classes, you can buy low cost index funds that own many individual REIT stocks, thus diversifying your risk among many companies in this business. Generally REITs pay out most of their income in the form of dividends, so they can be good for income. However, they can be quite volatile, so you must be prepared for that if you include them in your portfolio.
REITs are considered by many as one of the best asset classes to add diversification to your portfolio, since they often have low or even negative correlation to the broad stock market. For example, in the January 2000- early October 2002 bear market, the Vanguard REIT index fund actually increased in value by about 20% as compared to a decline of about 47% for the Vanguard Total Stock Index fund.
However, this isn't always the case; in the most recent bear market (from September 2008 until early March 2009), the Vanguard REIT index fund declined by about 65%, as compared to the Vanguard Total Stock Index fund which declined about 47% during the same period. The good news about REITs for the most recent period is that the Vanguard REIT fund has increased over 100% since the March 2009 lows, while the Total Stock index fund has increased about 70% (as of mid-January, 2010). The large decline followed by a large rebound illustrates the opportunity to benefit from disciplined rebalancing; i.e., if you had bought more shares of your REIT fund as the market plunged, you would've made big gains on the dollars added at the lower values as the fund rebounded. Of course this also would've been true for the Total Stock fund, but to a lesser extent. This is an example of higher volatility working in your favor.
I personally experienced the benefit of this approach (buying more shares when the value goes down) during the recent decline and rebound, with the REIT fund as well as others. Having decided to begin significantly increasing my allocation to my Vanguard REIT fund in late 2007, the more it plunged in value starting in late 2008, the more I added to it -- painful though it was at the time! As of mid January 2009, I have a small overall gain in this fund over the last 3 years, and a somewhat larger gain since I first started investing in the fund in August 2005. If I hadn't bought more shares when the share price plunged, I would've had a negative total return in my REIT fund at this point (perhaps about -25% not including dividends, but the dividends would've softened the blow).
Wrapping it Up
There are other asset classes that some advisors/authors recommend, like gold and other commodities. Some also recommend further splitting your international investments into small, large, growth and value. However, I think we've taken it far enough in this article.
To summarize, here are the various asset classes you might consider if you want to go further than a total U.S stock, total international and one or more bond funds:
- Total U.S. Stock (large blend)
- Total International
- Emerging Markets (if your total international fund doesn't include them)
- Large Value
- Small Blend
- Small Value