How Smart Has “Smart Beta” Been Lately?

Let’s start with the definition from Investopedia. Smart beta investing combines some of the efficiencies and cost savings characteristic of passive investing while mimicking active investing strategies that have the potential to produce excess risk-adjusted returns, also called “alpha.” In a nutshell:

Smart beta uses alternative index construction rules to traditional market capitalization-based indices.

Smart beta emphasizes capturing systematic investment factors in a rules-based and transparent way. To the extent that these factors can consistently provide alpha, they are capturing systematic inefficiencies in market-weighted pricing.

Smart beta strategies may use alternative weighting schemes such as volatility, liquidity, free-cash-flow, quality, value, small cap, and momentum. All of these strategies have exhibited superior risk-adjusted performance to the S&P 500 during at least one multi-year historical period, thus showing empirical evidence for potential alpha.

According to ETF Global Insight (“ETFGI”) research, smart beta funds now command more than $1.1 trillion in total cumulative assets worldwide.

Smart Beta ETFs are comprised of groups of stocks index providers have determined as most representative of one or more of the above factors. In addition, recent studies have provided data contending that equally weighing a portfolio is a source of systematic incremental alpha over market-cap weighting. Therefore, in addition to the other eight ETF representing time-tested smart beta factors, we include equally weighted versions of two benchmark index ETFs. Also, please there are multiple ETFs built to capture the same factor. In most cases where several potential choices existed, the ETF with the highest level of assets under management was selected. All of the ETFs included in this study have ValuEngine reports available upon demand. In addition to benchmark ETFs, IVV, iShares S&P 500 Index ETF and QQQ, Invesco QQQ Trust representing the Nasdaq-100, 10 ETFs have been included in the study. They include the following list.

QUAL – iShares MSCI USA Quality Factor ETF – tracks an index of US large- and mid-cap stocks, selected and weighted by high ROE, stable earnings growth and low debt/equity, relative to peers in each sector.

COWZ – Pacer U.S. Cash Cows 100 ETF – screens companies based on projected free cash flows and earnings for the next two fiscal years, then ranks according to their twelve-month free cash flow yield. Free cash flow is the money a company has left after paying operating and capital expenditures. This is generally considered a quality factor. Dividing it by price gives COWZ a value-factor component as well.

VIG – Vanguard Dividend Appreciation ETF – presents another way of looking at the quality factor. Although “Dividend” is in its name, VIG focuses on dividend growth, not high yield. It specifically selects US-listed firms that have increased their dividend payments for the past 10 years.

VYMVYM- VYM offers a diversified approach to high dividend yield in a low-cost wrapper. Firms are ranked by forecast dividends over the next 12 months, those in the top half are selected and market-cap weighted.

SPLV – Invesco S&P 500 Low Volatility ETF – SPLV tracks a volatility-weighted index of the 100 least-volatile stocks in the S&P 500.

MTUM – iShares MSCI USA Momentum Factor ETF- tracks an index of large- and mid-cap US equities, selected and weighted based on price appreciation over 6- and 12-month periods and low volatility over the past 3 years.

IWN –-iShares Russell 2000 Value ETF- tracks an index of US small-cap value stocks. The index selects value stocks with low price-to-book ratios and similar value metrics from a universe of stocks ranked 1001-3000 by market-cap.

IWO – iShares Russell 2000 Growth ETF – tracks an index of US small-cap growth stocks, selected from those ranked 1001-3000 by market-cap based on higher price-to-book ratios and higher forecasted growth.

RSP – Invesco Equal-Weighted S&P 500 ETF – simply takes all the stocks in the S&P 500 and weights them equally. We compare this with its cap-weighted alternative, IVV, iShares S&P 500 ETF.

QQQE – Invesco Equal-Weighted Nasdaq 100 ETF – QQQE equally weights the stocks in the Nasdaq 100, It is compared with QQQ, Invesco QQQ Trust, cap-weighting the top 100 Nasdaq-listed US non-financial sector stocks.

Observations

For the past 5 years, the dominant index by far year-by-year and on a compound basis has been a benchmark, the Nasdaq-100 as represented by QQQ ETF. That is, the smartest index to have been invested in from our sample was not a Beta at all but a popular market index.

Another way of saying this is that the dominant systematic factors in the market have been mega-cap size, membership in the technology and/or communications sectors and listing on the Nasdaq. The first factor turns 50 years of factor research and empirical findings on their heads. Sector dominance has generally been shown to be cyclical in generating alpha. Most counterintuitive and devoid of fundamental underpinnings is the concept of being listed on one particular US exchange as a “smart beta” factor.

Equal weighting has NOT provided better returns for the stocks in the Nasdaq-100 weighting than the slightly modified cap-weighting used for QQQ. As shown in the table above, QQQ has outperformed equally weighted QQQE by nearly 400 basis points (4.00%) annualized over the past five years. The graph below shows the growth of $100 invested in each fund since inception.

In 11-1/2 years $100 invested in QQQE grew to $465 while $100 invested in QQQ grew almost 40% more to $642. Referring back to the data table, one can read similar results for the ETFs based upon the S&P 500. Market-cap-weighted IVV outperformed equally weighted RSP substantially in the 3-month, year-to-date, one-year and 5-year rolling return periods.

COWZ, combining a quality factor, free cash flow, with a value component by dividing it by price has been the top-performing “smart beta” ETF in the study with a 5-year annualized return of 9.43%, beating IVV, the iShares S&P 500 ETF by 177 basis points (+1.77%) per year. In fact, it is the only “smart beta” ETF that outperformed IVV. COWZ also provided a higher dividend yield, 2.1% as compared with 1.6%, and went down only half as much as the S&P 500 during the past 3-month period.

The next two smart beta ETFs, the ones that underperformed benchmark ETF IVV by the least, are QUAL and VIG both capture the quality factor, albeit in different ways. QUAL outperformed VIG in each period. VIG makes nearly half of the difference back during the past 5 years with a higher dividend yield and a lower expense ratio.

The other “smart beta” ETFs provided negative alpha, another way of saying underperformance, during the past 5 years. The factors that did not work include low price volatility, high dividend income and price momentum. Again, somewhat counterintuitively, SPLV, representing low price volatility, underperformed IVV during the third quarter, losing nearly 9% as compared with 7%. Generally, low volatility, correlated strongly with low beta, would be expected to go down less when the market goes down.

The Russell 2000 ETFs, representing small cap US stocks between 1001 – 3000 as ranked by float-adjusted market capitalization, greatly underperformed their large cap counterparts in mirror-like fashion. Once again, growth outperformed value as IWO, the small cap growth ETF, outperformed IWN, the small cap value ETF in every period. The axioms often heard quoted in the marketplace, “Studies have shown that over time, value outperforms growth, small cap outperforms large cap and equal weighting outperforms market-cap-weighting” all worked in reverse throughout the past five years.

Let’s get back to this article’s opening question, “How smart have smart beta ETFs been lately?” The answer is that they have performed miserably, consistently underperforming their “naïve beta” counterparts. This is not to say that the long-term research is wrong or presented with distortions. Indeed, this chart shows clearly that starting in 2003 through mid-October of this year, equally weighted RSP has in fact outperformed market-cap-weighted IVV since its inception, a period of about 20 years but that cumulative outperformance narrowed somewhat during the past 5 years. Similarly, researchers can still find long-term support for very long time frames supporting the value vs. growth thesis and the small cap vs. large cap thesis. Conclusions can differ depending upon the time frame that is presented. Perhaps the engineers that created these “Smart Bet” indices based upon historical data were not all that smart when it came to constructing factor index ETFs smart enough to outperform in the future. As a 40-year quant, I have concluded that what a quant does best is to predict the past with ever greater precision. As we all know from fund disclaimers, past performance is not a guarantee of future performance. More By This Author:September Seasonality Struck Again In 2023
Combining Cyber-Analysis And ValuEngine Model Forecasts For E-Commerce Companies
Case Study Of A Ten-Year Actively Managed ETF