Toronto Team  

  • T416.203.0067
  • 1.866.242.0203
  • F416.203.0544
  • 8 Wellington Street East
    3rd Floor
  • Toronto, Ontario M5E 1C5

Dividend ETFs: Value ETFs in Disguise

September 10, 2012 - 5 comments

Let’s set the record straight once and for all:  Dividend ETFs are Value ETFs

In order to convince those non-believers, I’ll need to pull out all the stops and run a 3-factor regression analysis* on the monthly returns for some of the most popular Canadian dividend and value ETFs.  For the analysis, I’ll use the following ETFs:

Dividend ETFs

Value ETFs


1.    Market Coefficient (a.k.a. “beta”)
ETFs with a beta equal to 1.00 would have a similar risk to the overall Canadian stock market.  As we can see by the results below, the dividend ETFs (value ETFs) do have slightly less (more) risk than the overall market.  This is why investors generally conclude that dividend stocks are less risky.  The T-stat for each fund’s coefficient was higher than 2.0, indicating a high level of significance for the results.

2.    Value Coefficient
This is the most important result from the analysis.  A coefficient of +1.00 would indicate a deep value fund.  The dividend ETFs have significantly high value coefficients (even higher than the value ETFs).  In other words, the expected returns of all four ETFs are highly dependent on the returns of value stocks in general (relative to growth stocks).

3.    Small-Cap Coefficient
XDV, XCV, and CRQ all have slightly negative small-cap coefficients (most likely indicating a slight tilt towards larger companies).  This makes sense, as many dividend payers are bigger, more well-established companies.  The only exception is CDZ, with a positive and highly significant small-cap coefficient of 0.20.  During the financial crisis, S&P’s strict index methodology forced the ETF to boot out the largest banks and insurance companies (since they failed to increase dividends), replacing them with smaller dividend paying companies that still met their criteria.  As the regression results show, this has increased the small-cap risk of CDZ dramatically (and with only 58 holdings, this doesn’t seem like an ideal choice for riskier small-cap/value investing). 

For all of you die-hard dividend junkies out there, I’m certainly not disagreeing with your investment strategy (on the most part).  Dividend investing is just another form of value investing, and should therefore be expected to have higher returns.  My only advice is that dividend investors take the time to better understand the risks they are assuming when they invest in undiversified dividend ETFs.


*Methodology for Construction of Canadian Regression Factors:

Mkt-TBill= MSCI Canada IMI Index – DEX 30 Day T-Bill Index

HML = ½(MSCI Canada Small Cap Value Index + MSCI Canada Value Index) – ½(MSCI Canada Small Cap Growth Index + MSCI Canada Growth Index)

SMB = ⅓(MSCI Canada Small Cap Value Index + MSCI Canada Small Cap Index + MSCI Canada Small Cap Growth Index) - ⅓(MSCI Canada Value Index + MSCI Canada Index + MSCI Canada Growth Index)


By: Justin Bender with 5 comments.
  20/11/2014 2:30:29 PM
Justin Bender
@William - I don't recall there being any significant alphas when I ran this analysis back in 2012. I'm going to be putting together an updated factor analysis of Canadian dividend ETFs, so stay tuned!
  20/11/2014 10:24:22 AM
Justin - awesome work as always. Were there any significant alphas observed in the factor analyses? Don't expect, just wondering...
  24/05/2014 3:42:09 PM
Thanks for your reply. Could you clarify the term factor in the context of stocks? I'm only familiar with the term factor when it comes to bonds.
  21/04/2014 12:32:30 PM
Justin Bender
@Grant - a portion of the expected returns on dividend stocks may be attributable to the term factor (similar to one of the known risk factors in bonds). This can help dampen volatility when broad-market stocks are doing poorly and bonds are doing relatively well.
  12/04/2014 6:02:10 PM
Given that dividend stocks are value stocks, and value stocks have higher expected returns and therefore higher risk than the market, why is that dividend stocks are less risky than the market - at least XDV and CDZ? Is it just that people just like getting dividends so are less likely to sell them during a down turn thus reducing their volatilty?

 Security code