Last week, we introduced you to QSG Management. QSG was a made-up name, but a real firm that has been in business since the 1980s. We used the Bloodhound System to learn a lesson in one afternoon what has taken QSG years to figure out on their own – if they have even learned the lesson at all.
QSG spoke to their investors about earning signficant excess returns over the general market by investing in good, quality growth companies. The strategy smells of sweet home-spun goodness. However, it wasn’t fulfilling its promise. The strategy may sound good on paper, but wasn’t actually providing significant excess returns. In fact, it was underperforming the general market. The problem that we pointed out wasn’t a poor execution of their strategy, but rather a poor strategy being executed.
Last week, we recreated the investment strategy of QSG by creating a replica of their investment process. It’s not perfect, but it was pretty darn close. This week, I wanted to take a stab at analyzing the portfolio and trying to find common threads that led to its underperformance. This is a living blog. I am typing these words on the fly without knowing if I will come to any logical conclusion, but feel free to join me for the journey (or just skip to the end).
My first step was to look at QSG’s current top 10 holdings. In a portfolio of 25-35 names, the top ten is highly representational, especially in their case where it equated to 50% of assets.
When Bloodhound was created, our founders tasked the Bloodhound System with an artificial intelligence project. We have 1,400 parameters (fundamental and technical) for which to backtest. The combination of those factors ranges in the trillions. However, the System took some simple parameters and paired them with other simple parameters, logging the results in the meantime. At the end, the System stored slightly over one million strategies in a library. The benefit of that is that we can search that library with all kinds of criteria including performance attributes including risk & return, parameter composition, or holdings.
Out of the one million strategies, thirteen of them currently seven of QSG’s top ten holdings. Although each of them had different buy rules to build a candidate pool and varying returns, there was one dominant factor among all thirteen. They all used the same ranking factor – 10Yr Sales Growth. Although there were a number of factors used to identify QSG’s quality companies, the overriding factor was a long-dated revenue growth trend. Although QSG’s top holdings represented 35% of these strategies’ holdings today, the ranking method alone doesn’t dictate performance. If you recall, QSG’s three-year underperformance compared to the S&P 500 was almost 3%. However, these same 13 strategies with a common ranking method ranged in over/under-performance from -4.3% to +17.8%.
If we expand the search to include 50-stock portfolios (as opposed to 10- and 20-stock holdings) and require that eight of the top ten holdings be currently held, we get similar results. Five 50-stock computer-generated portfolios held eight of QSG’s top ten holdings. The common ranking factor among those five strategies was 5yr Sales Growth. It appears that revenue growth is an important component to QSG.
Beyond search the computer-generated library, I can also compare the QSG holdings to strategies I have built myself. A section of our Research Tab called Pulse ranks any stock against any of my holdings. Now, my strategy collection is a hodge-podge of different strategies created for this blog, my own personal interest, and/or inquiries from clients. Another user with a more clear-cut set of strategies may glean more information that I. However, I did note an interesting coincidence. A long while ago (August 2012 to be exact) when I was just learning the System, I build a relatively intricate strategy that I called, “High Quality.” Although it did little to narrow the universe of stocks into a smaller pool, it did have a relatively complicated ranking method that weighed debt ratios, EBITDA margins, Receivable Turnover, Sales Growth, Cash holdings, and PEG ratios.
I learned pretty quickly, that my qualification of “High Quality” didn’t perform too well. In fact, the 10-stock holding portfolio is negative in each of the last three years.
Of particular note, the more broad the portfolio, the better it performed. That is a condemnation of my ranking methodology. As I add more lower ranked companies, my returns improve. Whoops. That fact holds up over an extended period as seen by the 27-year average returns:
Three of QSG’s top ten holdings rank quite highly in that “High Quality,” but underperforming strategy.
Compared to other strategies I monitor, it is not surprising that QSG’s top ten holdings rank highly among strategies that focus on growth in cash flows. However, they rank poorly when cash flow growth is compared to earnings growth. They also rank poorly/highly in strategies that seek cheap/rich valuations. Some of the names rank high in strategies that track money flow and relative strength, suggesting they follow a bit a of momentum strategy.
We also ran a blog post on companies returns following periods of low and high CAPEX intensity. Although the QSG’s holding don’t rank considerably high in either category, they do have a better fit to the high CAPEX group than the low intensity group. Our analysis from that post suggests that CAPEX doesn’t predict returns as some suggest, but rather suggests companies that have performed well are lead into incremental investments which may be leading to future underperformance.
Our expectation is that QSG gets wrapped up in a “goodness” of a company without weighing enough on valuation. As many of us know, everything has a price and a good company doesn’t necessarily equate to a good investment. It appears that QSR may fail to weigh sentiment into the equation. If everyone knows its a good company, the chances of building a new investor base to support the stock valuation diminishes. As an example, QSG added a new holding per their last 13F. It became their second largest holding during the second quarter tallying about 7% of assets. It’s 5yr sales, earnings and cash flow growth were 33%, 36% and 35%, respectively. However, at the time the 13F was filed, there were six “Strong Buy,” nine “Buy,” two “Hold,” one “Underperform” and zero “Sell” ratings according to Briefing.com. By the end of the second quarter, that medical supplies company was trading a 25x cash flow, 30x earnings, 6x book value and 8x sales. All this from an already $20 billion equity capitalization.
Perhaps I am showing my value/contrarian-bias here, but that doesn’t ring like a great investment to me.