In past Mindfully Investing articles I’ve presented many reasons why mindful investors should heavily favor stocks over bonds, at least until bond yields climb well above their current levels. Further, it’s mindful to use low-cost index funds and moderately diversify the types of stocks in your portfolio because no one can predict which stock types will perform best in the future. However, I haven’t written much about how within-asset class diversification (holding various stock funds) differs from between-asset class diversification (holding a mix of stocks and bonds). Because diversification within and between asset classes are both called “diversification”, you might assume they produce similar results, but that’s usually not the case.
So, I was intrigued when I saw a tweet a few days ago implying that stock diversification can produce benefits similar to stock/bond diversification. The tweet led me to a blog post by Eric Nelson at Servo Wealth Management. While most of his post focuses on historical returns data showing that small-cap value stocks have outperformed the broader stock market, he goes on to say:
- Diversifying across stocks and bonds is a familiar approach for most; very few, on the other hand, know that just as much bang for your buck is possible by diversifying within the stock market.
Nelson presents this graph to support his conclusion.
The blue line in the graph shows rolling 10-year annualized returns for the total U.S. stock market minus the returns from a 5-year U.S. government bond. Positive values on the blue line indicate that U.S. stocks outperformed U.S. bonds for those 10 years. The green bars show the return difference between small-cap value stocks and the total U.S. stock market. Green bars with positive values indicate that small-cap value stocks outperformed the total stock market for those 10 years. Nelson points out that the relative see-saw performance of these three assets shows that the benefits of diversifying stock holdings with small-cap value stocks are similar to the benefits of stock/bond diversification. As you will see, my conclusion from this evidence is a bit different.
Stock Performance See-Saws
First, the see-saw performance in Nelson’s graph is typical for any diversified stock portfolio. I showed numerous examples in a recent post on stock diversification, and because Nelson is focused on small-cap value stocks, I picked one of my graphs showing the historical see-saw performance of small-cap stocks relative to large-cap stocks over rolling 10-year periods.¹
If you look at the green bars on Nelson’s graph from 1981 onward, they match pretty closely the pattern of ups and downs for small-cap returns on my graph. That’s not too surprising, but I also showed that this same see-saw pattern occurs when value, growth, or international stocks are added to a portfolio containing U.S. large-cap stocks. What does this pervasive see-saw performance within diversified stock portfolios reveal about the quality and characteristics of stock diversification?
Diversification Basics
As I’ve summarized before, most research points to two main potential benefits from diversification. One is the potential for higher returns and the other is the potential for lower risks (lower volatility in this case). The obvious problem with these touted diversification benefits is that they’re “potential”.
Boosting Returns – For example, if you add some small-cap value stocks to your stock portfolio, there’s no guarantee that they will generate better returns than the broader stock market in the next several years. Looking at Nelson’s graph, the current lackluster performance of small-cap value stocks could continue, it could get much worse (like the period from 1930 to 1940), or it could get better (like in the 1970s and 1980s). While you can hope for a sporadic returns boost from adding a new asset to your portfolio, it’s more likely that the see-saw performance of two or more assets will simply smooth out your portfolio’s returns over time. That’s because when one asset in a portfolio performs well, the other assets are performing relatively poorly by definition, which dilutes the portfolio’s overall returns.
Nelson emphasizes that small-cap value has outperformed the broader stock market 83% of the time going back to 1927. So, if that superior performance continues, I agree that adding any amount of small-cap value stocks to a broad stock portfolio will boost returns. However, attributing that returns boost to “diversification” is misleading. That’s like saying that all the ingredients in a bowl of chili combine to make it spicy when we know that only the spicy ingredients cause the burn. I have a question for those who diversify by picking the best historical performers: “If you’re so sure a particular asset will continue to outperform, why not just fill your portfolio with 100% of that asset?” The obvious answer is that everyone knows it’s impossible to predict the best performing asset for an uncertain future. We place bets across several assets because we can’t predict the future. That’s the real reason to diversify.
Reducing Risks – On the volatility side, many types of diversification have historically reduced volatility as compared to holding a single type of stock fund. This most obviously happens when a lower volatility asset is added to a high volatility portfolio, for example adding a bond fund to an all-stock portfolio. But more interestingly, adding moderate amounts of higher volatility assets can in some instances reduce the overall volatility of a portfolio. That’s because very volatile assets jump around so much that their movements often occur in opposition to other assets in the portfolio. However, I’ve noted before that a few percent difference in routine volatility is inconsequential to the typical long-term investor, and usually fails to cushion portfolio losses during market crashes. For example, most un-mindful investors will still be tempted to panic sell when their portfolio’s value plummets by “only” 35% instead of 40%.
Within- and Between-Asset Diversification
These diversification basics help us better refine our main question. In terms of smoothing returns and reducing volatility, to what extent does stock-only diversification give the same “bang for the buck” as stock/bond diversification? Nelson’s graph doesn’t directly address this question because it shows only the differences in returns between assets. Differences provide little information about the direction or magnitude of the individual asset’s returns in any given period.
For example, if you look at Nelson’s graph for the 10 years from 2000 to 2009, small-cap value stocks outperformed the broad stock market portfolio by 15% annualized. (It’s worth noting that Portfolio Visualizer reports a difference in this period of only 10%. So, Nelson’s calculation methods are a bit unclear. And although Nelson probably used Ken French’s small-cap value data, the reliability of small-cap data prior to about 1962 is questionable.) But consider the chart below from Portfolio Visualizer, which shows the returns for each year from 2000 to 2009, with small-cap value shown in red and the broader stock market (represented by the S&P 500) shown in blue.
In 7 of the 10 years, small-cap value stocks and the broader stock market moved in the same direction. Only in 2000, 2001, and 2007 did these two stock types move in opposite directions. Perhaps more importantly, small-cap value stocks clearly failed to provide meaningful diversification during the financial crises in 2008.
While looking at annual returns over one 10-year period is helpful, standard correlation analysis is a better method for quantifying the comovement (or lack thereof) of two assets. Correlations calculated over rolling periods (just like the rolling returns discussed above) paint a better picture of the diversification potential from various pairs of assets. So, here’s a graph of the correlation between U.S. small-cap value stocks and large-cap stocks on a 3-year rolling basis from Portfolio Visualizer.
A positive correlation on the vertical axis indicates that two assets move together more often than they move opposite each other. A positive value of 1 means that the two assets move in perfect lockstep. A negative correlation means the two assets move opposite each other more often than not. For the period available in Portfolio Visualizer, small-cap value stocks were highly correlated with large-cap stocks with correlation values ranging from slightly above 0.5 to greater than 0.9, with an overall average of 0.79.
Compare that to the correlation between large-cap U.S. stocks and a U.S. intermediate duration government bond fund (with an average duration of 5.2 years, which is similar to Nelson’s 5-year bond comparison).
Intermediate duration U.S. bonds have provided much less correlation with large-cap stocks ever since the turn of the century, with correlations ranging from 0 to less than -0.5 (average for the period of -0.12). Bonds move opposite to large-cap stocks much more often than small-cap value stocks do.
Some of you are probably wondering what these correlations tell us about the ability of stock versus stock/bond diversification to smooth returns and decrease portfolio volatility. The answer to that question depends on how much of the diversifying asset you include. For example, adding 20% small-cap value to a large-cap stock portfolio will generate entirely different results than adding 80%. Fortunately, there’s an easy way to look at a range of diversification percentages, which goes by the unhelpful name of “Efficient Frontier”. It’s just a cross plot of volatility versus returns for a series of portfolios with different mixes of two assets. Here’s the Portfolio Visualizer graph of portfolios containing varying amounts of small-cap value and large-cap stocks.
The blue dot nearest to the red “U.S. Stock Market” symbol represents a portfolio with only 5% small-cap value stocks added. The blue dot nearest the yellow “U.S. Small Cap Value” symbol represents a portfolio with 95% small-cap value stocks. Given small-cap value’s historical superior performance, it’s unsurprising that the more of it you add, the better the returns. And although adding small amounts of volatile assets can sometimes reduce a portfolio’s volatility, adding small-cap value stocks to a stock portfolio only increases volatility.
Now let’s compare the stock diversification return and volatility performance to a portfolio that mixes U.S. stocks and U.S. intermediate bonds. Here’s that graph.
Note that the horizontal scale on this graph ranges much lower than in the previous graph. At first glance, the shape of these two curves appears similar, but there are important differences. First, adding bonds to a stock portfolio consistently decreases returns and decreases volatility. Second, the low end of the curve shows that adding small amounts of stocks to a portfolio containing primarily bonds will increase returns with very little increase in volatility. Third, and perhaps most importantly, the range of volatility for stock/bond diversification (about 5% to 15%) is entirely below the range we just saw for stock diversification (about 15% to 18%). With more stock diversification, you’re almost always adding portfolio risk (as measured by volatility), with stock/bond diversification you’re almost always decreasing portfolio risk.
Although these graphs are helpful, there still a bit academic. Let’s look at it in more human terms by examining the impact of these risk variations in 2008 when the world of finance was in crisis and virtually every market was crashing. Let’s use Portfolio Visualizer to compare a portfolio of 50/50 U.S. stocks and small-cap value stocks to a 50/50 portfolio of U.S. stocks and intermediate bonds. To understand how differently the two diversification schemes react to a crisis, we can look at the volatility, worst year, and maximum drawdown of these portfolios in the 10 years from 2001 to 2010 as shown in the first two rows of this table. And let’s benchmark that against undiversified portfolios as shown in the last three rows of the table.
| Portfolio | Volatility (Std. Dev.) | Worst Year | Max. Drawdown |
| 50% U.S. Stocks/50% U.S. Small-cap Value | 16.8% | -32.9% | -55.2% |
| 50% U.S. Stocks/50% Intermediate U.S. T-Bonds | 9.9% | -22.8% | -33.7% |
| 100% U.S. Stocks | 15.4% | -37.0% | -51.0% |
| 100% U.S. Small-cap Value | 17.8% | -32.1% | -56.1% |
| 100% Intermediate U.S. T-Bonds | 5.8% | -4.3% | -10.7% |
The increased risk inherent to stock diversification (first row) was obvious to anyone with this type of portfolio in 2008. On the other hand, the lower risk of stock/bond diversification (second row) is shown by all three measures. While a few percent in maximum drawdown wouldn’t placate most investors, the more than 21% difference in maximum drawdown between these two diversified portfolios is emotionally significant.
Conclusions
You might conclude that mindful investors should embrace stock/bond diversification, but the opposite is true. The reasons are plain if we look at the returns performance for these same two diversified portfolios from 2001 to 2010:
- 50% All-Stocks/50% Small-cap Value – 12.7% annualized return
- 50% All-Stocks/50% Bonds – 9.3% annualized return
Mindful investors favor stock diversification exactly because it often produces higher returns in compensation for enduring higher volatility. Nelson is trying to convince investors to embrace stock diversification, and mindful investors should entirely agree. The only problem comes when Nelson tries to convince us that stock diversification offers the same “bang for the buck” as stock/bond diversification. For all the merits of stock diversification, claiming that it’s the same as stock/bond diversification is simply a bridge too far. With stock diversification, you have the potential to grow more bucks but with a greater chance of a much more damaging bang to your portfolio.
Because I largely agree with Nelson, I regret if this post comes across as a take-down piece. Nelson has given a great deal of thought to the merits of diversified stock portfolios and the historical benefits of owning some small-cap value stocks. I think he knows that stock diversification is fundamentally different from stock/bond diversification, but in his zeal to promote stock-diversification, he left out some important caveats.
1 – Nelson makes comparisons to the total U.S. stock market, not the S&P 500, which is a large-cap U.S. stock index. Nonetheless, the performance of the total stock market is nearly identical to the performance of the S&P 500 in almost any period you might want to examine. This just shows how much large-cap stocks dominate the returns of the total U.S. stock market on a capitalization-weighted basis.
