A Simple Small-Cap Value Screen
(Guest Commentary By Bill Rempel May 8, 2008)
Dear Subscribers and Readers,
For those who had wanted to learn more about individual stocks, the art of stock selection, and model-based trading/investing, it is again time to see what one of our regular guest commentators, Bill Rempel, has to say. Bill is a prolific writing on the stock market and individual stocks and is the author of a very active market blog at: http://billrempel.com (The Rempel Report).
In this commentary, Bill is going to introduce and discuss a simple small cap value screen based on P/E, P/B and P/S intra-industry comparisons. Both the historical performance and the drawdown statistics of this system look promising. This screen also offers something for the purely fundamental investor as it services to act as a very good initial filter for those who want to do further fundamental and qualitative research. This commentary is very timely, in light of our most recent commentary on small cap value stocks last weekend. Without further ado, following is a biography of Bill:
Bill Rempel (aka nodoodahs) is an active poster on the MarketThoughts forum as well as a few others around the web. Bill is a regular, monthly guest commentator on our website (see Putting Recent Volatility into Historical Perspective for his last guest commentary). Bill graduated from Caddo Magnet High School (a high school for nerds) back in 1985 and proceeded to learn the hard way when he drank his way out of a scholarship to Tulane later that year. After a few years of sweating for a living, he decided to go back to school, and graduated from LSU-Shreveport in 1995 with a Bachelors in Mathematics - all the while working the overnight shift stocking shelves in a grocery store.
Post-college, Bill has been in the P&C insurance industry as an actuary, product manager, and pricing manager. Bill and his wife Millie are amateur investors with a variety of holdings, but they prefer to buy and hold value investments. In typical "value" style, they live cheap, driving old cars and preferring to save or invest instead of buying fancy "stuff."
Disclaimer: This commentary is solely meant for education purposes and is not intended as investment advice. Please note that the opinions expressed in this commentary are those of the individual author and do not necessarily represent the opinion of MarketThoughts LLC or its management.
Long-time readers are aware that I started out primarily as a "value investor." What attracted me to value investing was the structure of the process, the research behind the stock selection, and the proven effectiveness of the approach as well as a large degree of natural stinginess and an ability to be patient while holding a position! Even though the pursuit of structured, researched mechanical trading has led me away from "value investing," I still keep experimenting with value-based strategies that can be mechanized easily, and today I'd like to present a simple small-cap value screen that could either be used in a mechanized way by a trader with sufficient capital, or be used as a starting point for a discretionary small-cap long-term "investor."
Over the years, I've experimented with a lot of different initial screening criteria, including valuation (FundaTechnical) metrics such as Price/Earnings and Price/Book, company fundamental metrics such as Return on Assets, Earnings Quality, and Debt/Equity, and technical metrics such as Insider Buying or 1-Year Total Return. My most recent experiments, in value screening and in other techniques, have been about pruning away different criteria and finding the bare minimum of "what works."
My latest experiment in "value" proposes that we evaluate stocks on a relative basis within their own industry, to avoid industry selection problems cause by the relative opaqueness of some business categories. For example, financials generally trade at some valuation "discount" due to market participants' doubts about what's in or on their books; industries go in and out of favor; cyclical industries vary in multiple based on market participants' views of where we are in the "economic cycle;" etc.
I will use three valuation metrics, and demand that the stock is cheaper than its industry group average in all three: Price/Earnings, Price/Book, and Price/Sales.
I will use four fundamental metrics, three of which are taken up by demanding the stock have positive Earnings over the last 12 months, have a positive Book Value in the most recent quarter, and have positive Sales/Revenues over the last 12 months. Additionally, I demand that the stock have a positive Return on Assets, and that its ROA be better than its industry group average.
Since this is intended to be a small-cap value screen, I will limit myself to exchange-traded stocks above $100 million (no micro-caps or over the counters) and sort all qualifiers Ascending by Market Capitalization, giving preference to the smallest companies first.
In Keelix terminology, it will look like this:
Keep :OR([SI Exchange]="N",[SI Exchange]="A",[SI Exchange]="M")
Keep :[SI Market Cap Q1]>100
Keep :[SI PE]>0
Keep :[SI Price/Book]>0
Keep :[SI Price/Sales]>0
Keep :[SI Return on assets 12m]>0
Keep :[SI PE]<[SI Ind. PE]
Keep :[SI Price/Book]<[SI Ind. Price/Book]
Keep :[SI Price/Sales]<[SI Ind. Price/Sales]
Keep :[SI Return on assets 12m]>[SI Ind. Return on assets 12m]
Sort Ascending [SI Market Cap Q1]
While one could use the screen as a starting point, to identify candidates for discretionary trades, I tested it as a mechanical process.
I tested intervals of 1 month, scanning and changing out stocks each time based on qualifications, always sorting by the smallest market cap. I tested holding counts of 10, 20, 50, 100, and the maximum possible number of stocks as well. Results do not include transaction costs or dividend payments, and many of these stocks do pay dividends.
Here's where some interesting rubber met the road.
Holding 10 by market cap means that stocks which appreciate gain in cap size, and may move out of the sort order; perhaps a better idea would be, if grabbing only 10 stocks, to hold them for a quarter (or even a year!) before rescreening and rebalancing! It would certainly help with transaction costs and position turnover.
Holding 20 by market cap presents the same problem in a slightly lesser degree, but also adds a smoothing effect from increased diversification one mitigates individual position risk by holding more stocks. Past 20 stocks, many retail investors don't have the capital to trade this system in a mechanized manner, because then transaction costs pile up, especially as a percentage of equity. The solution may be holding for a quarter, or a year, instead of a month.
I am not presenting those longer-hold tests of 10 and 20 stocks here they are left to the reader, or to a later post.
Holding 50 presented the best result in terms of CAGR (cumulative annualized growth rate). This might be an artifact of the high turnover in holding 10 or 20 stocks, however. Holding 100 or more lowered the CAGR, which is to be expected, if the long-term outperformance of small-cap stocks is any guide. CAGR went down again as market cap went out the window in my "hold the maximum number" test, which wound up carrying 298 stocks, on average, over the test period. Note that this method, which would be suitable for an institutional money manager, basically amounts to holding all stocks over $100 million that are better than their industry average in Price/Earnings, Price/Book, Price/Sales, and ROA. It amounts to giving zero preference by market cap. The lowest turnover was achieved with this method.
||Hold Period (Days)
||Turnover (Each Period)
The problems with turnover and some stunting of returns are directly due to the sorting by market capitalization in combination with smallish holding counts. I do believe there are better ways to accomplish both the sort and the concentration on small cap stocks, but I will address those in a later post.
Here are the equity curves represented by a hypothetical $1 put in various hypothetical funds, at the start date on 8/31/1997. The S&P 500 gain is shown for comparison. Transaction costs and dividends are not included in the results, so I'm specifically limiting this comparison to the 50, 100, and MAX holding counts, because these could be executed on an institutional scale with limited transactions "drag." Mechanized retail investors, pay attention, I promise I'll revisit the 10 and 20 hold tests at a later date.
Notice that from mid-1998 through early 2001, the various "value" funds would have dramatically UNDERPERFORMED the S&P 500. There are many different interpretations and lessons to be derived from this; keep in mind this was a period when the world's most famous value investor, a Mr. Warren E. Buffett, was considered to have been "out of step" and to have "lost his touch." For me, the primary lesson is that good FundaTechnical mechanical systems always work over the very long term, but don't always work over the short or even mildly long term, and this is a reason why Joel Greenblatt is OK giving away his "Magic Formula" and why mechanical systems never get their "edge" arbitrated away because most people would have abandoned this system during that timeframe.
Hey, does Bill Miller come to anyone's mind right now?
Here are the drawdown curves for the different hypothetical funds. Remember, a drawdown is when the system's in a losing period and the equity is below a prior peak.
Notice that the maximum drawdown for each fund was in late 1998 at about 30%, and it made another drawdown in late 2002 near the time of the market bottom. Also notice that the system was inordinately punished in the last few months, as value investors all took a good ol' Texas whupping compared to the S&P 500 since the October 2007 peak.
At this point, I'm satisfied from my knowledge of stock market "anomalies" and some limited backtesting that this stock screen has merits, and the only thing left to do is see how it might be used. This doesn't fit into my mechanical inclinations at the moment, but I'm thinking that it makes a fine and dandy first filter for discretionary value investors! I view the backtest results as a "proof of concept" small cap, cheap in its industry by multiple measures, and a better return on assets than its industry, equals better than average returns. It's a starting point, and it'll get more work in the future.
If you flash back with me to earlier posts on discretionary value investing, the screens are there to bring qualified candidates to us; it's our job to interview those candidates. We examine the balance sheets, cash flow statements, and income statements for signs of shenanigans; we read the filings and proxies for material risks and conflicts of interest; we try to understand the business model, whether or not the company has a "moat" or good prospects; all this before pulling the trigger. It is at once a more time-consuming and less-diversified endeavor than mechanical investing, but not necessarily a less-rewarding one.
As of the time I'm writing this post, here are some sample candidates:
Wow. Some of those really give meaning to the idea of a value investor holding his nose when he buys! No recommendations and no current positions in any of these, as always I fully disclose my positions on my blog, this is just a presentation of the simple small-cap value screen, some backtests on it, and some current candidates.