The Strengths and Weaknesses of the Major Value Metrics

Episode Summary

Executive Summary: The episode reviews the most common value metrics used in investing—price-to-book, price-to-earnings, price-to-sales, price-to-cash flow, and EV/EBITDA—explaining how each works, its strengths, weaknesses, and where it appears in Validia models. The hosts emphasize these ratios are screening tools, not true intrinsic value measures, and argue composites and ranking approaches can reduce factor-specific weaknesses and cyclicality.

Main Topics: Value ratios as screening tools, not intrinsic value (Priority: 5/5): The hosts stress that common valuation ratios are shorthand ways to find statistically cheap stocks, not direct estimates of what a company is worth. They are usually combined with quality or fundamental filters in real-world strategies. Price-to-book: widely used, widely criticized (Priority: 5/5): Price-to-book is presented as the most common academic and ETF value metric, prized for lower turnover but criticized for ignoring intangibles, profitability, and fair value of assets. Price-to-earnings: simple but accounting-sensitive (Priority: 5/5): PE is framed as intuitive and profitability-oriented, but highly dependent on how earnings are defined and vulnerable to accounting distortions and cyclicality. Price-to-sales and price-to-cash-flow: more robust, but incomplete (Priority: 4/5): Price-to-sales is harder to manipulate than earnings but ignores margins and profitability; price-to-cash-flow is more intuitive because cash is central to valuation, though it can also miss important accounting realities. EV/EBITDA and enterprise-value thinking (Priority: 5/5): EV/EBITDA is described as closer to how an acquirer values a business because it accounts for debt and cash, and it has performed well historically in O'Shaughnessy's tests. Composites and ranking methods (Priority: 4/5): The discussion explains that combining metrics via rankings can smooth factor cycles, reduce weaknesses of any single ratio, and potentially add quality exposure. Beware short-term factor judgment (Priority: 5/5): The hosts warn against declaring any valuation metric 'dead' based on a few years of underperformance; statistical noise and long factor cycles require decades of evidence.

Key Arguments: Valuation ratios are not used in isolation in Validia models; they are starting points that are typically paired with quality or fundamental tests. Price-to-book is popular in both academic research and ETF construction because it is a stock measure, tends to lower turnover, and has been central to classic value studies like Fama-French. Despite its popularity, price-to-book is criticized for ignoring intangible assets and profitability, making it arguably the weakest common value metric on balance. PE is easier to understand than some other measures, but its usefulness depends heavily on how earnings are defined (trailing, forward, or averaged) and can be distorted by accounting rules or cyclical earnings peaks. Price-to-sales is difficult to manipulate because revenue is harder to massage than earnings, but it fails to account for profit margins and can unfairly compare high-margin tech firms to low-margin businesses. Price-to-cash-flow is conceptually closest to discounted cash flow valuation because owners care about future cash generation, though cash flow can sometimes misrepresent economic reality compared with earnings. EV/EBITDA adjusts for capital structure and approximates an acquirer’s view of the business; it has shown strong historical performance in O'Shaughnessy’s testing and is often favored over simpler ratios. Composite value approaches can improve robustness by blending metrics that capture different parts of the financial statements and by reducing dependence on any single ratio’s performance cycle. A different explanation for composite benefits is that metrics like price-to-book embed negative quality, and combining factors can remove that weakness rather than simply diversify factor timing. Investors should avoid drawing conclusions from short periods of underperformance because factor efficacy can vary over long cycles and statistical significance may require far more data than a decade.

Data Points: Price-to-book coverage in value ETFs: Almost every value ETF uses book value or includes it as part of the process - Jack notes book value is the most used ratio in ETF value strategies, citing Eric Balchunas' ETF analysis. Lowest 20% of stocks: Bottom 20% by price-to-book - Describes the first screen in the Piotrowski strategy before applying F-score fundamental tests. PE example: $20 stock / $2 earnings = 10 PE - Used to illustrate how the PE ratio is calculated. Lower turnover: Price-to-book tends to produce lower turnover than earnings- or cash flow-based ratios - Because book value changes less frequently than flow-based measures. Lowest 1% to 2% tile: Bottom 1%-2% by EV/EBITDA - Describes how the acquirer’s multiple model selects candidates using ranking rather than fixed thresholds. Historical test start date: 1956 - O'Shaughnessy’s EV/EBITDA testing is described as going back to 1956. Statistical significance horizon: Longer than a decade; likely longer than an investing lifetime - Jack references Corey Hofstein’s work on how long it takes to determine a factor is truly dead.

Pivotal Quotes: "these metrics are shorthand ways to try to find stocks that are cheap" — Justin Carboneau / Jack Forehand: Summarizing the purpose of valuation ratios versus intrinsic value models. "you can't even use a decade because ... the amount of time you need to say with statistical significance that one of these factors is dead is probably longer than any of our investing lifetimes" — Jack Forehand: Warning against declaring valuation factors obsolete based on short-term underperformance. "when we're creating a composite of all these value factors, what we do is we take each individual factor, we rank every single company in our database using that factor, and then we combine those rankings" — Jack Forehand: Explaining how a value composite is constructed in practice.

Implications: Listeners should treat valuation ratios as imperfect screens, not truth machines. The best practice is to combine multiple metrics with quality checks, use ranking-based composites thoughtfully, and judge factors over decades rather than recent cycles.