Finding The Right ‘Smart Beta’ ETF

January 28, 2015

This article is part of a regular series of thought leadership pieces from some of the more influential ETF strategists in the money management industry. Today's article is by John Eckstein, chief investment officer and director of research at Astor Investment Management. For the latest smart beta news and analysis, visit out smart beta ETFs channel.

"Smart beta" is as hot as it is poorly defined. It seems everyone wants to squeeze a bit more return out of their portfolio by switching to a different core equity holding. 

And why not? Smart beta promises the safety of an index along with the return boost of a systematic trading strategy. This article is the first of three to look at the record of some of the larger and more important smart-beta strategies currently on offer. In this part one, I examine how smart beta ETFs test their indexes before going live and examine the results of the indexes behind dividend focused ETFs. Future articles will examine other regions of the smart beta terrain.  

I'm going to define smart beta by taking both words seriously. 

I take "beta" to mean that the strategy has to be a modification of some major index. And in fact, that is the proposition on offer: Sell some of your S&P 500 Index and buy something else to replace it. The "smart" means abandoning the cap-weighted standard and finding some other sort of way to weigh constituents. A smart way, ideally.

This gives us some clarity, but leaves an interesting collection of what ETF.com classifies as alpha-seeking strategies on the sideline for now. These funds do have a mechanical trading rule, but they are too far from a baseline index to qualify under my definition. 

These funds are only about 10 percent of the size of the funds I classify as smart beta, but they could be important source of noncorrelated returns. So, we'll return in to these in a future article.

We can divide the smart-beta funds that remain first into dividend-focused funds and the rest. The dividend-focused funds have by far been the most successful, with about 75 percent of the $100 billion in smart-beta ETF assets invested in large-cap smart-beta funds managing more than $300 million, according to the latest data from ETF.com. 

Watch Your Back(test)

Before we can finally take a look at the performance of some of these funds, we must stop to consider the track record. 

Smart-beta funds select a set of rules to do their investing. They will define in advance the universe of stocks to be considered and how that large list will be narrowed down, and how the weights will ultimately be determined. 

With a strict mathematical rule, the strategies can be simulated in the past before the fund launched or before the index rules were chosen. We call these simulations "backtests." If there is an iron rule of backtests, it's that every one that sees the light of day looks good—in simulation, at least. The live indexes of the typical large dividend-focused smart-beta funds go back to about 2006, while the indexes have been simulated back to about 2001. 

If you divide each index's history into before and after its rules were published, on average, the alpha—that is, the pure return of the strategy—decreases. Meanwhile, its beta—the return that is attributable to the stock market—increases. I interpret this as evidence suggesting that a long history of real-time operation is a point in a strategy's favor. 

 

Dividend-Focused Funds

Dividend-focused smart-beta funds tend to pick a subset of the index with the most constant dividend record. Some of the funds are further screened for financial stability or a record of increases in dividends.

I need to mention that while high dividends have intuitive appeal, there are two caveats. 

First, in theory, dividend-paying companies may simply be those without any better ideas to invest their money. Another way of looking at that is that you'd probably rather have a company you own invest in a project with 15 percent return than have it pay 3 percent back to you. On the other hand, some think that dividends force management to be more disciplined. 

Second, because of tax law, companies also buy back stock instead of paying dividends. In fact, in 2013, the "buyback yield" exceeded the dividend yield. (See this paper by NYU's Aswath Damodaran for more information).  It's not clear how this shift will affect returns of dividend-focused funds in the future. 

Despite my comments above, I am going to use the fund's published indexes, minus the current ETF fees, instead of looking at the ETFs. That will give us a longer period to analyze. In general, the managers of these funds track their indexes as closely as possible. I'll look at two time periods: a longer one from 2001-2014; and a post-financial crisis from 2010-2014. 

I'm looking at the bigger funds, roughly $300 million or more. The table below shows the compounded annualized growth rate and the annualized return, divided by the standard deviation to give a sense of the index's risk-adjusted returns. The worst drawdown—the  maximum loss from peak—is also reported.

 

  Compund
Annual
Growth
Rate
Annualized
Risk-Adjusted
Returns
Worst
Draw-
down
  Compund
Annual
Growth
Rate
Annualized
Risk-Adjusted
Returns
  Index
Live
  2001-2014   2010-2014    
VIG 5.5% 0.44 -41%   13.7% 1.11   2006-03
DVY 8.5% 0.56 -57%   16.7% 1.41   2003-03
SDY 9.1% 0.64 -50%   15.4% 1.24   2005-11
SCHD 10.5% 0.78 -45%   16.7% 1.43   2003-12
SDOG 11.7% 0.65 -53%   17.8% 1.36   2012-05
PEY 6.1% 0.34 -68%   16.3% 1.38   2005-11
NOBL 9.8% 0.72 -45%   17.6% 1.44   2005-05
                 
SPY 5.0% 0.33 -51%   15.0% 1.08   1954-01

Dividend-focused ETFs, performance of index tracked by funds, adjusted for fees

2001-2014. Sources: Bloomberg, ETF.com, Astor calculations

 

Examining Performance

So, how have these indexes done over time?   

For the seven indexes that have been extended back to 2001, all outperformed the S&P 500. In addition, all seven outperformed the S&P 500 on a risk-adjusted basis. However, when we look at returns through the lens of the worst drawdown, another measure of risk, we see the results are more ambiguous. 

While these results are good, for a more cautious interpretation of dividend focused funds with different data, see Larry Swedroe's article on ETF.com from last summer, distinguishing between high dividend and value strategies.  

Also note that the funds that simulate the best have been calculated in real time for the shortest amount of time. Even in the more recent period, when the S&P 500 has done quite well, the high-dividend funds have outperformed on both an absolute and risk-adjusted basis.

Note that the average correlation of the indexes to each other is about 0.92, and the average correlation to the S&P 500 is 0.91. In other words, not much diversification benefit to your core equity portfolio here. To earn a place in your portfolio, you have to believe these indexes will continue to produce superior risk-adjusted returns.


Astor Investment Management is a money manager with an active and economically grounded approach to asset allocation. We believe that investment opportunities arise based on the ability to identify fundamental trends and changes in the economy. We build portfolios of ETFs appropriate for our analysis of the business and monetary policy cycles. For more information, see www.astorim.com; for our blog, see www.astorinsights.com.

 

Find your next ETF

CLEAR FILTER