Momentum, Quality And Low Volatility: Continuing The Quest For Smarter Beta
Summary In November I introduced a smart beta portfolio based on MSCI’s indexes for quality, momentum and low volatility. The semi-annual rebalancing of those indexes is complete. I review the previous six-month performance and determine the components of the rebalanced MQLV portfolio. In early November I proposed the idea of using the iShares smart beta ETF portfolios as a filter for building one’s own risk-premia portfolio ( A Quest for the Smartest Beta ). I started from three ETFs, each indexed to a single factor: Low Volatility, Momentum and Quality. iShares MSCI USA Minimum Volatility ETF (NYSEARCA: USMV ) iShares MSCI USA Momentum Factor ETF (NYSEARCA: MTUM ) iShares MSCI USA Quality Factor ETF (NYSEARCA: QUAL ) Taken together, these three ETFs make a solid holding as seen in this table showing results of an equal weighted portfolio of the three ETFs vs. the S&P 500 since the inception of QUAL, the youngest of the three, in August 2013. (click to enlarge) Starting from the premise that each of the ETFs is selecting for a single “smart-beta” factor I wanted to look at the intersection of the three funds. I asked if there were overlapping positions in all three ETFs. I compared their full sets of holdings looking for that overlap. There were 14 funds shared by all three. I reasoned that since each of the 14 passed the MSCI filters for low-volatility, momentum and quality, it could be worth looking at a portfolio comprising all 14, in effect, a portfolio located at the intersection of Quality, Momentum and Low Volatility. June through November Results The 14 stocks from the end of May rebalance are: Arch Capital Group Ltd (NASDAQ: ACGL ) Accenture PLC (NYSE: ACN ) Axis Capital Holdings Ltd (NYSE: AXS ) Chubb Corp (NYSE: CB ) Chipotle Mexican Grill Inc. (NYSE: CMG ) Home Depot Inc. (NYSE: HD ) Eli Lilly (NYSE: LLY ) Nike Inc. Class B (NYSE: NKE ) O’Reilly Automotive Inc. (NASDAQ: ORLY ) Reynolds American Inc. (NYSE: RAI ) Starbucks Corp (NASDAQ: SBUX ) Sigma Aldrich Corp (NASDAQ: SIAL ) Visa Inc. Class A (NYSE: V ) W.R. Berkley Corp (NYSE: WRB ) Each of the ETFs is rebalanced to a revised index twice annually, on the last business days of May and November. So, when I looked at the portfolio, let’s call it MQLV , it had a five-month record from its “inception” on the last business day of May. It had performed well. For the five months from June 1 to Nov 1, it turned in a CAGR of 41.0% vs SPY’s -1.30%. Now that the full cycle is complete we can update performance at the close of the six-month holding period. It performed thusly: (click to enlarge) That is a quite impressive performance record. In a market environment where the S&P 500 index could only muster a 1.74% total return, MQLV chalked up nearly 19%. Sharpe (2.21) and Sortino (7.29) ratios are at rarely seen levels. Pretty good evidence that there may well be something to this idea. Not in any way definitive, of course; it is, after all, a single cycle. But those results are surely saying “Hey, look over here.” Rebalancing for December through May Now that MSCI has rebalanced the indexes, I let’s have a look at the changes. The current overlap for the three funds has moved from 14 to 18 stocks. Eleven remain from the previous list. There are seven new entries, and three have dropped off. The additions are: Costco Wholesale Corp (NASDAQ: COST ) Henry Schein Inc (NASDAQ: HSIC ) Lockheed Martin Corp (NYSE: LMT ) Mcdonalds Corp (NYSE: MCD ) Public Storage REIT (NYSE: PSA ) Travelers Companies Inc (NYSE: TRV ) Ulta Salon Cosmetics & Fragrance I (NASDAQ: ULTA ) And the deletions: Chipotle Mexican Grill Inc. Reynolds American Inc. Sigma Aldrich Corp CMG is no longer included in MTUM’s holdings but remains in USMV and QUAL. RAI was dropped from QUAL; it remains in USMV and MTUM. SIAL was acquired. The sector mix is dominated by Consumer Discretionary and Financials which account for 12 of the 18 positions. (click to enlarge) If we combine these 18 positions into an equal-weighted portfolio, the portfolio metrics are as follows: (click to enlarge) (from investspy.com based on one-year’s data) One-year performance for these 18 is outstanding, having beaten SPY 27.7% to 3.5% for the year. This is, of course, no indication of what the portfolio will do over the next six months between now and the next rebalance, but it does auger well for success. And, let’s not forget, 11 of these holdings were included in the previous iteration which trounced SPY handily. Here is a correlation matrix for the holdings. (click to enlarge) Running the portfolio through Portfolio Visualizer’s four-factor analysis produces the following regressions. Once again, it’s based on one-year’s data. (click to enlarge) As commenters pointed out in discussing the November article, there is little exposure here to size, all but three of the size exposures are negative. Several suggested that I should include the value factor. I argued that value was inherent in some of the selection criteria used by USMV and QUAL, so adding an ETF like the iShares MSCI USA Value Factor ETF (NYSEARCA: VLUE ) would be redundant. That point of view was confirmed to a large extent by including the VLUE and the iShares MSCI USA Size Factor ETF (NYSEARCA: SIZE ) portfolios in the analysis as a follow-up ( Expanding the Smart Beta Filter: Does It Help? ). Now, from the results of this regression analysis of the Fama-French factors, we can see that value exposure is, in fact, fairly high. This result confirms my sense that value was being addressed at least partially, even though it is not a specific factor for any of the three source ETFs. HSIC, LLY, LMT, SBUX are negative for value, but the rest are positive or neutral. Unsurprisingly, momentum exposure–the only factor specifically selected for by a source ETF–is high; only LLY is negative here. Given the extraordinary success of the June through November record I am excited to see how the rebalanced portfolio performs. At 18 positions this is a fairly large commitment for an outright investment, but it could well be worth some serious thought. To me, the concept appears sound and the track record, limited though it may be, is supportive. Is it actionable? I’d like to think so, but the hard evidence, however impressive, is sketchy. So any action taken would be largely based on an appreciation for the conceptual basis of the strategy. I’ll be keeping this updated as we move forward.