This column is part of a new collection of our “Structure Matters” series of interviews with leading ETF and index industry figures. They are conducted by Dan Weiskopf, a portfolio manager at New York-based Access ETF Solutions LLC. In today’s piece, Weiskopf interviews David Brown of Sabrient Systems, an index provider for both the Guggenheim Defensive Equity ETF (DEF | B-42) and the Guggenheim Insider Sentiment ETF (NFO C-55) as well as the Direxion All Cap Insider Sentiment ETF (KNOW | B-83).
Dan Weiskopf: David, your firm seems to have a theme going with your focus on insider buying and sentiment. What are the different aspects to your screening process for the sentiment factor, and how are the two ETFs [NFO and KNOW] that run on your Insider/Analyst model different in how they use this factor? What statistical aspects to insider buying make it a good predictive factor?
David Brown: Thanks for taking an interest in the Sabrient approach to thematic multifactor modeling. In the case of our Insider/Analyst model, I had long observed that insider accumulation appeared to have some predictive qualities, and of course, there is a history of academic research that has shown such relationships.
Moreover, isolating just the open-market purchases from the total Form 4 filings further improves the predictive power of insider-buying data. In addition, our sentiment studies found that there is not only an initial positive impact on a stock price after recent positive earnings revisions from Wall Street analysts, but a longer-term upward drift that continues to persist for a while thereafter.
But when we combined these factors, we discovered some interesting synergies that created even stronger predictiveness. In particular, we look at insider buying activity, i.e., both the number of corporate officers, directors and major shareholders who are buying on the open market over the past few months, and the magnitude of each executive’s increase in beneficial ownership.
We also look at Wall Street analyst upgrades, i.e., both the number of analysts who have issued positive earnings revisions over the past three months and the magnitude of each analyst’s EPS [earnings-per-share] increase. For example, did it increase from $0.50 per share to $0.52 or to $0.75?
Both groups of people are essentially insiders in a broad sense, because they are more or less “in the know” due to their intimate knowledge of the company and its industry. In effect, postive analyst sentiment can serve as a confirmation of insider optimism.
Weiskopf: Insider buying can be an indication of a management’s relatively short-term outlook. What can investors gain from current insider-buying patterns across sectors and market-cap sizes?
Brown: In all of Sabrient’s models, a bottom-up aggregation of scores for individual stocks can create relative rankings among sectors—or any stock basket, for that matter—including market caps or ETFs. This can provide insights into which sectors are displaying, for example, relatively strong positive sentiment and therefore might be poised to outperform.
Weiskopf: Your sector weightings across Guggenheim’s NFO and Direxion’s KNOW are substantially different, except for an overweighting in consumer cyclicals. What factors have led to these ETFs having different weighting?
Brown: Two key differences between the two Sabrient indexes underlying those ETFs are the limitation on sector concentrations and turnover. NFO enforces strict limits of 20 percent from any given business sector and 10 percent from any given subindustry. KNOW has no limits at all. KNOW can essentially take larger sector bets, so to speak, whereas NFO ensures broad diversification.
In addition, NFO has a 25 percent turnover limitation on each quarterly rebalance to reduce transaction costs, while KNOW has no turnover limitation at all and it rebalances monthly, which can lead to faster rotation in sector weightings.
Weiskopf: Can you speak about the other factors that make up your process in the indexes underlying KNOW and NFO? I don’t see a lot of overlap in the holdings. How does the screening process work beyond the insider buying?
Brown: The two underlying indexes both select from a range of capitalizations from small to large, and both employ the same basic underlying Insider/Analyst model to identify 100 stocks. However, there are significant differences in how we arrive at the final portfolios.
For NFO, we start with a broad universe of over 4,000 eligible stocks that we track in our database, and the Insider/Analyst model is run to identify the 100 highest-ranked stocks, subject to the sector and turnover limitations. All 100 stocks are equally weighted. Overall, the process is pretty straightforward.
On the other hand, Direxion’s KNOW selection process is more complex and active. It essentially combines aspects of both Insider Sentiment NFO and Defensive Equity DEF.
Our Defensive Equity Index employs a proprietary model that ranks how well a given stock holds up on particularly weak market days, i.e., an investor sentiment metric that also translates into relatively low volatility, as well as our Earnings Quality Rank (EQR), which is a model we developed with Sabrient subsidiary Gradient Analytics to measure the relative aggressiveness of a company’s accounting practices.
For KNOW, we start by selecting only from the S&P 1500, but we eliminate the bottom 20 percent of stocks as ranked by our EQR, leaving about 1,200 eligible stocks. We then run the Insider/Analyst model to identify the 100 highest-ranked stocks for the final portfolio—again, with no limits on sector concentrations or turnover.
Next, we apply our Defensive Equity model to rank these 100 stocks according to their tendency to outperform during periods of market weakness, and each stock’s relative ranking is used to determine a modified exponential weighting. As a result, the KNOW portfolio seeks to provide a strong combination of earnings quality, insider buying, analyst upgrades and defensive investor sentiment.
As you can see, KNOW is more complex, active, concentrated and unrestrictive in its search for the most compelling stocks for the coming month. Although some investors prefer such an aggressive approach, others prefer the process underlying NFO, as it is more conservative regarding turnover, transaction costs and diversification.
Dan Weiskopf: David, thank you for the time and the explanation of how sentiment can be used as an investment factor.
Chart courtesy of StockCharts.com
Dan Weiskopf is a portfolio manager at Access ETF Solutions LLC, whose third-party ETF strategies are offered through IPI Wealth Management, Inc. (IPI). IPI is an SEC-registered investment advisor, with its principal office located at 226 W. Eldorado St., Decatur, IL 62522, 217-425-6340. Access ETF Solutions LLC is not affiliated with IPI. Readers are advised to read the full transcript of the interview, including disclosures at http://accessetfsolutions.com, or to contact Dan Weiskopf at 212 628-4882. At the time of the intgerview, the manager may directly or indirectly hold a position in the Direxion Insider Sentiment ETF (KNOW) ETF. This interview should be viewed as an educational piece and not a recommendation.