This article is a continuation of my look at Dividend Stock ETFs and will focus on asset and portfolio allocation considerations. This will include calculating the betas of each ETF and evaluating their impact on an existing portfolio. While beta provides one approach to measuring systematic risk, it does not cover all risks in owning an ETF. Some investors view beta as a useful measure and others do not. Beta is a construct from theoretical finance and is part of the capital asset pricing model.
This article will use the same set of ETFs from Part 2 which are listed below:
|Ticker||Name||Family||Recent Price||TTM Yield|
|SDY||SPDR S&P Dividend Aristocrats ETF||SPDR||73.06||3.9%|
|SPHD||PowerShares S&P 500 High Dividend Portfolio||PowerShares||28.64||3.6%|
|HDV||iShares High Dividend ETF||iShares||69.78||3.2%|
|FDL||First Trust Morningstar Dividend Leaders Index Fund||First Trust||22.08||3.1%|
|DVY||iShares Dow Jones Select Dividend Index Fund||iShares||71.99||3.0%|
|VYM||Vanguard High Dividend Yield ETF||Vanguard||61.99||2.8%|
|SCHD||Schwab US Dividend Equity ETF||Schwab||36.39||2.5%|
|DLN||WisdomTree Large Cap Dividend Fund||WisdomTree||66.13||2.5%|
|DTD||WisdomTree Total Dividend Fund||WisdomTree||66.93||2.4%|
|VIG||Vanguard Dividend Appreciation ETF||Vanguard||74.97||1.9%|
|SPY||SPDR S&P 500 Trust ETF||SPDR||186.29||1.8%|
Source: Yahoo!Finance, Company Websites, Author calculations
Comparing the Betas of Dividend Stock ETFs
These ETFs offer varying dividend yields and P/E ratios as seen in Part 2. However, in making a selection, one should also consider different risk and volatility measures. As noted earlier, one measure is beta, which is the ratio of the covariance of a security to that of the market relative to the variance of that given security. This math simplifies down to beta being the product of the correlation between the security and the market and the ratio of the security's volatility to the market's volatility. So in the simplest terms, there is a correlation part - how does this security track the market - and a relative volatility part - is this more volatile or less volatile than the market as a whole. I've historically used SPY as a measure of the market which is pretty reasonably for U.S. based investors due to home market bias.
The following tables shows the construction of beta of each ETF over a 36 month period:
|Ticker||Correlation 36 months||Volatility 36 months||Volatility Ratio to SPY||Implied Beta|
Source: Monthly split and dividend adjusted prices from Yahoo Finance, Author calculations. SPHD, SCHD, and HDV do not have 36 month histories yet. Implied beta is the correlation times the volatility ratio.
One can see that FDL and DVY have much lower betas than the market in general. Furthermore, all the ETFs, except QQQ, were measurably lower than the market. As noted in Part 2, FDL had a lower P/E and a much lower growth outlook. However, given its more stable base and lower correlation to the market, it would be relatively more attractive. Beta is also time dependent. The following table looks at 24 month calculation for comparison. This will also add back in SCHD and HDV.
|Ticker||Correlation 24 months||Volatility 24 months||Volatility Ratio to SPY||24 Month Beta||36 Month Beta|
Source: Monthly split and dividend adjusted prices from Yahoo Finance, Author calculations. SPHD does not have a 24 month histories yet. Implied beta is the correlation times the volatility ratio.
This table shows that the betas do shift depending on time frame (which incidentally is one of the weaknesses of technical analysis); however, rank order remains similar and the shift was a uniform increase.
Measuring the Impact on a Portfolio
Another way to think about adding a new security to your portfolio is what type of impact it will have on portfolio's volatility. Are you truly lowering your systematic risk or are you increasing it. There may be other risk and return considerations, but those are for another article. The following table looks the impact of adding each of these ETFs to a 100% SPY portfolio to create a 50/50 portfolio.
|Ticker||Correlation||Security Volatility||New Portfolio Volatility||Volatility Impact|
Source: Monthly split and dividend adjusted prices from Yahoo!Finance, Author calculations. SPHD does not have a 24 month histories yet. Implied beta is the correlation times the volatility ratio.
As hypothesized, FDL offers the largest reduction in portfolio volatility. Also perhaps surprising is that adding QQQ creates only a slight increase. A larger increase would not have been surprising.
Checking Historical Performance
The following table shows the performance of the ETFs over various time frames.
|Ticker||1 year return||2 Year Return||3 Year Return||5 Year Return|
Source: Monthly split and dividend adjusted prices from Yahoo Finance, Author calculations.
This shows QQQ to be the best over the past year - so are we entering another Tech bubble and has the allure of dividends started to wear off? SPY follows QQQ which is not surprising if you think investors are shifting from value to growth. As noted in Part 2, the lower growth ETFs are posting lower returns in the past year, which is also not surprising.
Contrary to what I suspected, the dividend stock ETFs do not have the same beta as the market. Furthermore, there is reasonable variation among the ETFs, that it should be a consideration when making an investment. For those interested in lower volatility and current income - FDL looks more attractive. However, it is not entirely surprising that the higher yield, higher payout dividends stock ETFs tend to have lower growth, lower volatility, and lower returns (at least recently). In this group I would put FDL, HDV, and DVY. I would also hypothesize that SPHD is also in this group, which had a 60% payout ratio and low historical and projected growth.
The other group of dividend stock ETFs tends to have lower yields (but still about market) and higher growth. These ETFs also tend to have higher betas and larger volatility. Part 4 of this series will put Parts 2 and Parts 3 together to see if there are some discrepancies between value and risk and growth with a goal of identifying the best ETF option. While simple grouping approaches is a good way to distinguish, the reality is that these ETFS lie along a continuum for any given dimension.
Disclaimer: This article is for informational and educational purposes only and shall not be construed to constitute investment advice. Nothing contained herein shall constitute a solicitation, recommendation or endorsement to buy or sell any security.