Six Special Value Stats
4/27/2015
If you regularly read our stories on our Quadrix stockrating system, you know that Value is the most predictive of the six category scores used to derive the Overall score.
See below for more on the effectiveness of the Value score itself. This story drills down to some of the individual valuation ratios that make up the Value score. In particular, we focus on these six:
• Enterprise value/EBITDA (enterprise value is stockmarket value plus the value of debt minus net cash. EBITDA is earnings before interest, taxes, depreciation, and amortization).
• Price/book (book is the accounting value of a company's assets minus the value of its liabilities).
• Price/cash flow (in this shorthand version, cash flow is income plus noncash depreciation and amortization).
• Price/free cash flow (free cash flow is operating cash flow minus capital expenditures and dividends).
• Price/sales.
• Price/sales divided by the fiveyear median price/sales ratio.
Why these six ratios out of the 21 we use to derive the Value score? Because they have the most predictive power in the broad S&P 1500 Index since 1994.
Of the six, price/free cash flow delivered the biggest bang — the cheapest onefifth of S&P 1500 stocks based on this ratio managed 12month total returns an average of 5.0% higher than the average stock. No other value statistic came close to that excess return. However, investors shouldn't rely on price/free cash flow alone.
Portfolios constructed based on price/free cash flow tend to deliver more volatile returns than those based on other valuation ratios. In other words, the effectiveness of price/free cash flow varies greatly over time. In addition, a lot of companies don't report free cash flow, and many that do have negative values for it, which renders the price/free cash flow ratio useless. Only 607 stocks in the S&P 1500 Index have price/free cash flow ratios above zero and below 35 — and that's about 150 more stocks than the monthly average since 1994.
Other valuation ratios don't face the same issues as price/free cash flow, but they all have their pros and cons; none offers a truly representative picture of a stock's valuation. We prefer to consider more than one ratio — preferably more than two or three — even though valuation ratios are highly correlated with each other, as shown in the table below.
MEASURING CORRELATION


The table below presents correlations for the excess returns delivered by portfolios consisting of topscoring stocks based on six valuation ratios. For example, the enterprise value/EBITDA ratio is more highly correlated with price/cash flow (0.83) than with price/book (0.65). A correlation of 1.0 means returns matched perfectly; each of the statistics has a 1.0 correlation with itself.


 Correlation Between Outperformance Of Top Quintile Based On: 


Enterprise Value/ EBITDA 
Price/ Book 
Price/ Cash Flow 
Price/Free Cash Flow 
Price/ Sales 
Price/ Sales To 5Yr. Avg. 

Enterprise Value/EBITDA 
1.00 
0.65 
0.83 
0.58 
0.79 
0.54 
Price/Book 
— 
1.00 
0.60 
0.62 
0.86 
0.87 
Price/Cash Flow 
— 
— 
1.00 
0.52 
0.73 
0.48 
Price/Free Cash Flow 
— 
— 
— 
1.00 
0.71 
0.54 
Price/Sales 
— 
— 
— 
— 
1.00 
0.79 
Price/Sales To 5Yr. Avg. 
— 
— 
— 
— 
— 
1.00 
In a stockmarket context, correlation measures the tendency of one investment to move with another. A correlation of 1.0 means investments move in lockstep, while a correlation of zero implies no linear relationship. Excess returns of portfolios assembled based on the six most effective value scores are fairly well correlated.
We rely on more than 20 individual statistics to compute the Value score. While the scores are highly correlated, some work better than others, and some work better in some sectors than in others. Since we don't know whether historical patterns will repeat, we can't predict which Value variables will work best in the future. By using a variety of Value variables, we reduce our risk of getting burned by one or two metrics losing efficacy.
The table below presents 12 Arated stocks that score above 50 in at least four of the six key valuation statistics, as well as above 70 in Value and above 80 Overall. Three are reviewed below:
No way can we call Foot Locker ($60; FL) a value stock. However, we do see the stock as a value. Its Quadrix Value score of 72 tops the average of 67 for apparel retailers in the S&P 1500 Index. Foot Locker doesn't exceed 69 in Quadrix for any individual value metrics, but it also doesn't have many truly pricey ratios, and as such qualifies as a solid value, particularly relative to expectations. At 15 times expected yearahead earnings, Foot Locker trades 22% below the average apparel retailer.
In recent years, people in the U.S. and Europe have adopted more active lifestyles and purchased shoes to match, a trend that seems to have staying power. Over the last year, Foot Locker has grown sales 10%, pershare profits 25%, and operating cash flow 34%. The consensus projects pershareprofit growth of 10% in fiscal 2016 ending January and 11% in fiscal 2017. Both estimates have risen in the last 60 days and still seem somewhat conservative. A steady stream of new and innovative products should help keep demand strong, while store remodeling continues to boost sales productivity. Foot Locker is a Focus List Buy and a LongTerm Buy.
The biotechnology industry isn't known for discount valuations. Biotech stocks average Value scores of 27, and Gilead Sciences' ($104; GILD) rank of 86 tops the nextcheapest rival by more than 30 points. Gilead trades at 13 times trailing earnings and 13 times operating cash flow, both ratios well below respective industry averages of 30 and 22. Of course, Gilead didn't get that cheap without reason. Investors have legitimate concerns about the stock, most notably worries about how deeply it must discount the hepatitis C drugs that have driven its incredible growth.
However, analysts expect profit growth of 19% this year and 8% next year, targets that reflect a slowdown in revenue growth from hepatitis drugs — and don't fully take into account Gilead's robust portfolio of drugs under development for HIV and liver disease. HIV medications accounted for $10 billion in sales last year (41% of the total), and a potential new blockbuster could be approved by midyear. In February, Gilead declared its firstever quarterly dividend of $0.43 per share, payable in the June quarter and equating to an indicated yield of 1.7%. Gilead, expected to declare Marchquarter earnings April 30, is a LongTerm Buy.
Travelers ($102; TRV) earns a Value score of 92 and exceeds 80 in three of the six valuations ratios most effective at predicting outperformance. Property & casualty insurance stocks as a group are cheap, averaging Value scores of 81, but Travelers stands out from the pack because of its other scores. The stock earns a 95 Overall and is one of only three stocks in the industry with Momentum and Quality scores above 70 as well as a Value score above 90.
In the March quarter, Travelers earned $2.53 per share, down 14% and $0.01 below the consensus. Net written premiums fell 1% to $5.90 billion. A combination of lower investment income and a gain on an insurancelaw change in the yearago quarter suggested profits would fall, but the results were still somewhat disappointing. However, retention rates, premium prices, and newbusiness volumes rose, reflecting decent underlying operating momentum. Travelers raised its quarterly dividend 11% to $0.61 per share payable June 30, marking the 11th consecutive annual hike, and authorized enough share repurchases ($5 billion) to shrink the stock count 14%. Travelers remains a Buy and a LongTerm Buy.

Value works well, but shouldn't stand alone
In rolling 12month periods since 1994, top Value scorers outperformed the average stock in the S&P 1500 Index by an average of 2.8%. Value also delivered the highest excess returns over the last one, three, five, 10, and 15 years. In fact, top Value scorers delivered higher average returns than our flagship Overall score during every period.
Given those numbers, you might ask, "Why not just use the Value score and ignore everything else?"
The answer revolves around volatility. Top Value scorers averaged 2.8% outperformance since 1994 and top Overall scorers managed excess returns of 2.3%. However, the Value score's returns were less consistent.
Portfolios driven by Overall score were more likely to outperform than Value portfolios, and their outperformance wasn't as widely dispersed. Based on standard deviation, a measure of how numbers are distributed around the average, the outperformance of Value portfolios was more than twice as volatile as Overall portfolios. Most important, we don't know if Value will remain so effective. The Overall score is a more diversified bet on what will work on Wall Street in the future.
We measure the relationship between return and volatility by dividing standard deviation into average outperformance. Using that metric, top Overall scorers deliver a return/risk ratio of 0.46, well above the Value score's 0.27. As the table above illustrates, the Value score's return/risk ratio dwarfs those of the other category scores. Using six categories that don't all outperform at once creates the power of the Overall score, which is designed to work regardless of which types of stocks are in favor.
Quadrix return comparison
 Quadrix Scores 


Overall 
Momen tum 
Value 
Quality 
Fin'l Str. 
Earns. Ests. 
Perfor mance 

Average Outperformance (%) 
2.3 
0.4 
2.8 
0.3 
(0.9) 
(7.2) 
(0.3) 
Standard Deviation Of Outperformance (%) 
5.0 
6.3 
10.2 
6.4 
6.7 
14.8 
11.3 
Return/Risk 
0.46 
0.06 
0.27 
0.04 
(0.13) 
(0.49) 
(0.02) 
Note: All numbers are based on rolling 12month returns since 1994, except for Earnings Estimates, for which our data starts in 2004. 