“The logic behind the magic formula is that it pays up for quality”
-Alpha Architect
The team at Alpha Architect dissected Joel Greenblatt’s Magic Formula which equally weights two factors in its construction. Put simply these metrics are 1) ROC (typically identified as a “quality” factor) and 2) EBIT/TEV (a “value” factor).
While they weren’t able to replicate Greenblatt’s results in backtesting, the conclusion Alpha Architect made from separating the MF methodology into single-factor portfolios was that the EBIT/TEV value factor contributed to almost all of the annualized CAGR outperformance over a 40-year period.
The returns of the EBIT/TEV factor portfolio came on a lower max drawdown, similar downside deviation, and a much higher percentage of winning months against the ROC-centric portfolio over a 40-year period.
Research more recently has argued building a portfolio of stocks that screen as statistically attractive on “quality” metrics can be a trap for buying into mean reversion. Sustainable “moats” are probably difficult to screen for.
Back to the Global Macro Grind…
We’re not going down the rabbit hole on the right and wrong ways to systematically capture “quality.”
Rather we’ll discuss some obvious dangers with factor buzzwords like “growth” and “value”. These style identifiers have gained popularity in recent years, particularly when marketed as low-cost alternatives to actively managed portfolios.
Factor and Smart Beta vehicles are popping up faster than we can track. We’re asked quite a bit about what “growth” and “value” mean to us. The simple context we communicate is that growth and value are defined by sectors and style basket performance in different growth and inflation environments. Factor exposures are defined through our Growth, Inflation, Policy Model Framework. We use style indices and ETFs that have already been created if they fit the framework. We then employ market-based metrics for trend confirmation and risk management.
The danger we see in the factor and smart-beta marketing push for assets right now at the Retail and Advisor level is in the potential failure to look passed stylistic titles. What exposure are you trying to capture and what does it provide you in the context of a total portfolio?
For a real-time example of factor strategy idiosyncrasies, Consumer Discretionary and Information Technology have historically been among the best performing S&P 500 sectors in a growth accelerating environment as we define it. However, many popular “value” strategies that may also chase low volatility momentum, particularly the ones that have more flexible sector constraints, are heavy into Consumer Discretionary and Retail right now. Growth and Value are often treated as mutually exclusive.
Alpha Architect’s Quantitative Value strategy defines its universe and constructs its strategy with some of the following rules:
- Top 60th percentile in the NYSE
- Throw out Financials/ADRs/REITs/ETFs
- Extensive forensic accounting and earnings quality/manipulation screens among other granular checks to further shrink its universe.
- The strategy is then constructed from the value decile of that universe based on the EBIT/TEV metric. The portfolio is rebalanced quarterly based on the same criteria.
The end result is a portfolio that has a much smaller # of names relative to most closet index-hugging vehicles (~40 currently) with much more aggressive sector constraints (less-constrained). In short, the strategy is looking to capture the long-term value premium in a single factor and is openly willing to take the sector and time-horizon risk to do it.
Remember the most widely-televised trend 6 months ago that Amazon was going to strangle brick and mortar retail equity holders and creditors?
That scenario hasn’t gotten much press lately…
Through 1H of 2017, the SPDR S&P Retail ETF underperformed the S&P 500 by 1580 bps. Its cumulative return of +10.9% since the beginning of 2017 is still well below the S&P 500’s return of 26.7%, but it made up some serious ground in Q4 (+21.6% over the last 3 months alone).
Quantitative Value (QVAL) has a 34% tilt to Retail currently. The next closest industry weighting is Media at 11%. A quick scan of the buyside holders list of the most heavily weighted stocks in QVAL reveals the usual suspects who run strategies with similar principles.
Over the last year, QVAL has crushed the Russell 1000 Value Index and is almost keeping pace with the Russell 1000 Growth Index which has been a core “growth accelerating” exposure of ours since we introduced our Q1 2017 themes:
1Yr Performance of Index ETFs:
- Quantitative Value (QVAL): +34%
- Russell 1000 Growth (IWF): +37%
- Russell 1000 Value (IWD): +17%
The Russell 1000 “Value” index is constructed with Price-to-Book and Forward Earnings Growth Rate factors and ~70% of the Russell 1000 index members are included under the “value” or “growth” umbrellas.
Exposure to the Russell 1000 Value index takes on a tremendously higher amount of interest rate sensitivity than the aforementioned “Value” ETF which holds no Financials or REITs. Banks are the most heavily weighted industry group in the Russell 1000 Value Index at 15%. The bottom line is that these two “value” factor exposures are a world apart.
The goal in calling out short-term performance isn’t to blow smoke up anyone’s rear-end. Different strategies work better in different environments. The point is to exemplify how convoluted the barrage of smart-beta identifiers can be.
It’s dangerous to throw the dart at a generic style tilt because it has a lower fee. Take a hard look. Although you already knew that if you have any longevity in this industry.
Our immediate-term Global Macro Risk Ranges (with intermediate-term TREND views in brackets) are now:
UST 10yr Yield 2.51-2.68% (bullish)
SPX 2 (bullish)
RUT 1 (bullish)
VIX 9.14-12.67 (bearish)
USD 89.15-91.63 (bearish)
Oil (WTI) 63.06-64.93 (bullish)
Good Luck Out There,
Ben Ryan
Macro Analyst
