Using Quantitative Strategies to Pick Winning Mid-Cap Stocks
by Charles Rotblut, CFA and Brian Peery
Brian Peery is a co-portfolio manager for Hennessy Funds, and co-manages the Hennessy Focus 30 fund (HFTFX). I spoke with him about incorporating stocks screens into the portfolio management process.
Charles Rotblut (CR): You are using a quantitative strategy to identify stocks for the Hennessy Focus 30 fund (HFTFX), correct?
Brian Peery (BP): Correct.
We use the Compustat database, which is composed of roughly 10,000 public companies. Our whole goal is to design model portfolios that will outperform the relative benchmark. We want to accomplish that by taking a lower-risk profile, so the model has to outperform the benchmarks, but with lower risk. The one that we designed and implemented for mid caps is our Focus 30 fund.
On an annual basis, we run our parameters. Our first step is to segment out companies that are between $1 billion and $10 billion in market capitalization, which are what we define as mid caps. We also remove ADRs (American depositary receipts), which are foreign companies that are traded on a domestic exchange, because we only want U.S. companies. We found that when we put ADRs in the portfolio, we increased the fund’s volatility, but not necessarily the returns. The nice thing about the mid-cap space is that it’s kind of the quintessential sweet spot of the market, where companies are large enough to have a defensible market position, but small enough to be nimble and hopefully grow their business quicker than some of the large-cap companies. In our point of view, the mid-cap segment tends to be a really kind of underappreciated market.
The next screen looks for a price-to-sales ratio of below 1.5. People ask why we incorporate the price-to-sales ratio in our models versus utilizing the price-earnings ratio or the price-to-book-value ratio. What we’ve found historically is that the price-to-sales ratio is a better indicator of a company’s true value. If you think about it, when you do your tax returns at the end of the year, you want to take as many legitimate deductions as you possibly can to get your income as low as possible. That means you take write-offs this year on capital gains if you can, push them forward or do whatever you need to do to get your taxable income to where you want it to be within the constraints of the law. And that’s exactly what companies do. They want to show analysts a nice, smooth earnings growth line, where it suggests everything is rosy. Alternatively, when you employ revenue and the price-to-sales ratio as a factor in your model, you are using a measure that companies cannot adjust. If a company is manipulating its revenue or price-to-sales ratio, then they have to be cooking the books, like Enron. Fortunately though, we’ve never experienced anything like that in our portfolios. Our relative value guide directs us to purchase companies whose price-to-sales ratio is less than 1.5, which means we won’t pay more than $1.50 for $1 in sales.
Our next step is to review the firm’s rolling 12-month earnings. Quarters are great; you can see an increase on a quarterly basis. But we want to see it on a longer-term basis. So we look at that 12-month time period to make sure that the revenue growth we typically see in the top line is actually being transferred to the shareholder in the form of higher earnings. It doesn’t have to be a positive earnings number—it can go from negative $0.30 to negative $0.10; the number just has to be going in the right direction.
Our final screen searches for relative market capitalization, which is just a way of saying we want the stock price to be higher. So, we’ll look at the relative strength over a three-month period, six-month period and a 12-month period. And if it’s positive in all three periods, then it makes our final cut. It’s kind of like the theory of inertia, when something in motion tends to stay in motion. For us, it’s kind of like going to a baseball game with a 3–0 lead. It doesn’t necessarily mean you’re going to win at the end of the day, but it’s sure nice having that tailwind behind you. [Editor’s note: relative strength compares the return of a security to the return of a benchmark; it tells you whether the security is performing better or worse than its benchmark.)
So once the companies are past all of those screens, we construct our portfolio by ranking the remaining stocks by their 12-month price appreciation. Based on that, we select the top 30 and allocate equally to each of them. We purchase 3.3% of the assets in the portfolio in each of the top 30 names.
We choose to include 30 stocks because when you’re above 17 to 19 names, you diversify away a significant portion of your risk. So for us, we wanted to have a portfolio diverse enough where we thought that people could get exposure to the market, but we didn’t want to be an index fund where you are buying 2,600 stocks. We wanted our particular picks to have some impact on performance.
Also, we don’t keep a significant cash position; we keep less than 5% because we don’t get paid to manage cash. Shareholders want us to invest for them; that’s our job.
Once we have our 30 stocks and 3.3% allocated to each, we’ll hold them for roughly a year. We ideally like to hold them for a year and a day because that gives the company some time to work out. When we rebalance the portfolio, if a stock continues to meet our criteria, we’ll continue to hold that position, but we’ll reduce the weighting back to the original 3.3%. We take some profits off the table and reduce the risks, so we’re not allocating 7% or 8% of our portfolio to an individual security. If a stock no longer meets our criteria, then we’ll sell it and replace it with one that does. We typically have around a 90% turnover in the fund on an annual basis.
Everything we do, at least in terms of this fund, is laid out in the prospectus. It’s all in black and white, and nothing changes. For us to change part of the methodology requires a shareholder vote.
Last 3 Yrs
Last 5 Yrs
|Hennessy Focus 30 (HFTFX)||5.9||4.0||19.4||1.2||0.0||1.36|
|Average of Mid-Cap Stock Funds||7.0||-3.1||16.9||0.6||0.3||1.07|
|Source: AAII’s Quarterly Low-Load Mutual Fund Update, January and July 2012. Data as of June 30, 2012.|
CR: After you run the screen and identify the stocks that meet all this criteria, you then rank them by 12-month return? And you just buy the top 30 from the screen?
CR: Any concern about things that don’t show up in a stock screen?
BP: We do vet the top line. We go through everything, there’s all sorts of subsets as we’re doing these models; we triple-check, we get data from multiple sources. We check earnings, we check any pending litigation, if there’s merger and acquisition activity going on. If we find a company we selected was falsifying income statements or doing anything that would have precluded their original selection, it would be removed from the portfolio. We haven’t had that issue, but it’s something that we monitor on a daily basis.
I was a fundamental analyst for a long time, and I’ve done the management tours and talked to a lot of companies throughout my career. What I find is that if you simply let your stock selection process dictate your picks, and if you let discipline guide you, I think you’ll do better over a long time period because you don’t get caught up in the emotions, the fads or the trends that are so prevalent in the stock market. We’re not trying to pick sectors and say, “OK, we’re going to shift into this because we think that’s where it’s a hot place to be.” That’s when people tend to get swayed and chase after returns, and most of those investors experience poor returns. So if you can maintain that discipline, I think you’ll outperform many investors over time.
That’s really what our formulas do; they don’t let us inject our personal opinions about what stocks to buy. That’s one of the great things about the process.
CR: We have a lot of screens on the AAII website, over 60. A common question I’m asked by AAII members is whether they should simply buy everything that passes a screen. But you’re saying the screen’s results are simply the suggested portfolio, and you perform due diligence before you actually buy the stocks just to make sure there’s not a boogeyman in the closet.
BP: Yes and no. There’s a lot of other things we look at—liquidity and some other underlying factors and such. Once you determine exactly what your parameters are, though, I think you really do have to stay within those boundaries unless some factor comes along that causes you to change the original conclusion.
CR: I know you said you don’t look at price-earnings ratios because earnings can be manipulated, and, obviously, the price-to-book-value ratio depends on what a company puts on the balance sheet and how they’re valuing the intangibles. What about price relative to cash flow, how do you look at that? It’s a little bit harder to manipulate if you’re not using EBITDA (earnings before interest, taxes, depreciation and amortization).
BP: We actually use the price-to-cash-flow ratio in some of our other models. And what we’ve found, interestingly enough, is if you take Morningstar’s boxes and draw the nine boxes, what works in the value box doesn’t necessarily work in the growth box. And it doesn’t necessarily work between large cap and small cap. So within those realms, each model has different factors that make those stocks attractive within those boxes. So our large-growth portfolio looks completely different than our large-value portfolio in the selection process. The selection process for our Focus 30 fund uses completely different parameters than that for our large-growth fund. [Editor’s note: Morningstar’s equity style box can be seen in this issue on page 13.]
CR: And then with earnings, you’re just looking at the last 12 months. Correct?
BP: Well, it’s actually the last two years. We use the rolling 12 months as our basis, but we go back to the year prior to that.
CR: You’re not worried about the long return, you just want to see the recent growth?
BP: Yes. I want to see the last two years to make sure they’re going in the right direction.
Even during the down time in ’07, ’08 and kind of into ’09, it really helps to point out companies that are doing the right thing. While the revenue might not be growing as steeply, if you can get ahead on the cost curve and start really cutting costs effectively to drive earnings in a down season, those are companies you could really want to own.
CR: As long as a stock passes the screen and nothing really bad appears during the due diligence, you’re going to buy the stock? Even if the margin trend might not be positive, you’re going to stick to the basic criteria?
BP: Yes. There are a thousand things that you can look at. There are all sorts of trading strategies you can do and try and implement and tweak things. But what we’ve found was that if we can keep it kind of simple and easily understandable, it makes it a great deal easier if something is broken, you can actually find out where it is and fix it.
It’s all too easy to pick and choose and manipulate the data that you want. For instance, I could make a fund that says, “take The Wall Street Journal every Thursday and pick every 40th stock or 43rd stock.” I bet you I could find some pattern that beats the market, but there’s has to be some basis. There has to be a reason why it happens.
CR: For relative strength, you just look for positive numbers, but is there a preference for ranking stocks based on one-year versus six-month relative strength?
BP: We’ve looked at all of the parameters: three-month, six-month, two-week, 12-week, 20-day, 50-day, 120-day, whatever the rolling averages are. What we found was that one year tends to be that sweet spot for what I’ll call momentum, because it’s long enough to give it time to work out, but it’s short enough that the actual momentum is sustainable.
When you go out to 15 months or 18 months, that momentum starts to drop off. When you shorten the duration, you wind up with a higher turnover and subsequently higher costs, which can substantially reduce your realized returns. And same holds true for the rebalance period.
CR: What about concentration—having certain industries dominate the screen?
BP: We’re limited to what subsectors we can invest in. As a mutual fund, I cannot have a portfolio concentrated more than 25% in one subsector. So, if one of the 30 qualifying stocks would cause the portfolio to become overly concentrated in, say, retail apparel, I would simply go to the next stock; I’d take the 31st on the list.
The fund has roughly 30% concentration in consumer discretionary, but there are nine subsectors within the consumer discretionary area that we own currently. A Family Dollar Stores (FDO) is nothing like a Tractor Supply Co. (TSCO); there’s enough diversification within that consumer discretionary space that we don’t hold stocks that all do the same thing. They’re widely spread out within that area.
CR: For an investor looking to follow a similar process to picking stocks, where they’re relying on a quantitative strategy to manage their portfolio, is there anything they should think about? Any suggestions?
BP: One thing that we tell everybody is to do an annual rebalance. If you started your portfolio in 1995 with 10% technology stocks, 60% large-cap stocks and 30% bonds and you left it just like that, by 2000 you’d probably have closer to 40% allocated to technology stocks. That’s not what your original risk profile looked like. Take the money off the table, move it and rebalance your portfolio so your allocation looks like your original risk profile. That’s the best thing you can possibly do. That way, you maintain your true risk profile and you don’t get overextended with potential volatility.
Brian Peery is a co-portfolio manager for Hennessy Funds and co-manages the Focus 30 Fund (HFTFX).