By Justin Carbonneau (@jjcarbonneau)
Interested in chatting with Justin – let him know.
Suppose you wanted to find the value stocks in today’s market using common value ratios. Which would you use and why? Would you lean more heavily on the price-to-book ratio, which is largely used in academic testing and originates out of the world of Graham & Dodd? Or would you rather use something like the price-to-earnings ratio, which is more common and uses a firm’s earnings. Or maybe since earnings can be noisy you’d prefer the price-to-sales. Or maybe instead you’d like to evaluate companies like a private buyer might and you want to use enterprise value-to-operating earnings, which helps adjust for a company’s capital structure. Or maybe you’re a fan of cash flow, so price-to-cash flow or free cash flow yield is best.
Picking a valuation ratio can be confusing. But the more appropriate question may be, why pick one at all? Whatever your preferred ratio may be, you may benefit more from a merger of value ratios vs. just using one.
But first a little context. Here at Validea, we track 22 distinct stock selection models. The majority of them are value strategies or have a value basis. For instance, in our model based on Joel Greenblatt’s Magic Formula, the main value metric is the earnings yield. In our Joseph Piotroski based strategy, it’s the book-to-market (the inverse of the price-to-book). In our Ken Fisher method based his book Super Stocks it’s the price-to-sales ratio. And in our Ben Graham model it’s a combination of the P/E and P/B. It’d be difficult to look at these models to determine what value metric might be the best one to use for the long run since they all vary.
But fortunately for you and me, someone has already done the research and legwork to determine what mix of value metrics should be used if you want to put the odds in your favor in a value portfolio.
What Works on Wall Street
James O’Shaughnessy, founder of O’Shaughnessy Asset Management, published the 4th edition of What Works on Wall Street in 2011. Taken from the book’s back cover — “What Works on Wall Street, Fourth Edition, examines actual historic performance figures to show you which strategies have provided investors with the best returns, which should be avoided at all costs, and why.” The updated version of the book contained data from the mid-1960s through 2009 and also new and improved investment strategies.
In the book, O’Shaughnessy presents a treasure trove of long term risk and return data on specific factors as well as combinations of factors. In the table below, which is taken from chapter 16 (The Value of Value Factors), we can see how a number of value related factors and screens have performed over time. The table shows what $10,000 investment made in 1964 would be worth in 2009 when invested in these various value portfolios focusing on specific factors.
If you look at the data, you can see how superior value stocks are compared to the anti-value stocks. You can see, for instance, how much better the low P/E stocks performed compared to the high P/E – a 5.53% annual return compared to a 16.25% return. At the bottom of the table, you will see three composites (Value Comp 1, Value Comp 2 & Value Comp 3) – all three of these composites produce a return higher than any of the individual factors in and of themselves.
source: What Works on Wall Street, Fourth Edition: The Classic Guide to the Best-Performing Investment Strategies of All Time
Merging Value Factors
The Value Factor can be deployed as a “pure play” on value or it can be combined with other metrics like Shareholder Yield or Buyback Yield to help improve the risk-adjusted results. O’Shaughnessy’s prior versions of What Works on Wall Street pointed to the price-to-sales ratio as being the best value metric, but the idea of combining multiple value factors together came as a result of a research paper O’Shaughnessy had read that combined price-to-sales with price-to-earnings ratios.
In the table below, you will see the three different Value Composite models along with the criteria in each. As O’Shaughnessy outlines in book, when you look at these factors over time the “best” performing factor changes. In one period it may be price-to-sales and in another period it may be EBITDA/EV, so the combined Value Factor model manages this jockeying by scoring stocks through at least five or six (in the case of VC2 and VC3 models) distinct criteria. And you’ll notice the balance sheet (P/B), income statement (P/S and P/E) and cash flow (EBITDA/EV and P/CF) are all represented by various factors.
|Value Factor One (VC1)||Value Factor Two (VC2)||Value Factor Three (VC3)|
|Pure Play Value Factors||Price-to-book|
|Enhancements||Shareholder Yield||Buyback Yield|
Implementing this can be a little bit tricky, as it requires all investable stocks to be scored and ranked and then for a cumulative ranking to be calculated based on all factors. We’ve captured the Value Composite Two strategy on Validea and so we do a lot of the heavy lifting for those looking to utilize the model in their own investing approach. Since the fourth edition was published, O’Shaughnessy’s firm has talked extensively about the limitations of the price-to-book ratio. As a result, the P/B is no longer used in their value composite based on our understanding. However, when we implement investment models we try to stay as close to the original published strategy as possible so we continue to use it. The move away from using the P/B, however, presents a good example of how value models can change and evolve over time.
They say that diversification is one of the only free lunches when it comes to investing. Usually, when you hear that it relates to diversifying among asset classes, but as O’Shaughnessy’s research shows, you can diversify across factors as well, even within the value category. And this approach gives us a superior long term result when compared to focusing on just one specific way to uncover value. The biggest takeaway might be there is no one superior value factor. When analyzing and searching for value stocks investors would be best served by looking at multiple factors like those listed above.
Photo: Copyright: 123rf.com / lightwise