Rafael Resendes On Intrinsic Value, Economic Margin and How Some Value Models Have Lost Their Way

Episode Summary

Executive Summary: The episode centers on Rafael Resendis’s critique of traditional quantitative value investing and his firm’s alternative intrinsic-value framework. He argues that book-to-price and related “cheapness” metrics became distorted by factor-hunting, low-quality evidence, and misread signals around growth, investment, and intangibles. He then explains Applied Finance Group’s live, 25+ year model built on economic margin, reinvestment, competition, and risk, and how it guides stock selection, portfolio construction, and occasional market calls.

Main Topics: Critique of traditional quantitative value investing (Priority: 5/5): Resendis argues that classic value factors like book-to-price have lost their original identity and now function more as cheapness screens than true value measures. He says the literature has morphed into a factor hunt rather than a coherent theory of intrinsic worth. Evidence quality and out-of-sample testing (Priority: 5/5): He stresses that much of the academic evidence behind value factors is backward-looking and subject to look-ahead bias, data snooping, and other weaknesses. He favors live, forward-tested data over post hoc historical observations. Investment factor and growth misinterpretation (Priority: 5/5): Resendis disputes the idea that higher investment should predict lower returns, arguing that the key distinction is whether a company earns above or below its cost of capital. He says models that penalize growth miss wealth-compounding compounders. Applied Finance Group’s intrinsic value framework (Priority: 5/5): He details a proprietary valuation system built around economic margin, reinvestment, economic profit horizon, and risk adjustments. The model capitalizes R&D and leases, adjusts for inflation, and aims to estimate true intrinsic value rather than accounting book value. Five-factor model and factor subsumption (Priority: 4/5): The firm’s five-factor model adds intrinsic value, financing yield, and leverage to the market and size factors. Resendis says these variables absorb much of the explanatory power of common academic anomalies such as HML, momentum, volatility, and investment. Portfolio construction, turnover, and sell discipline (Priority: 4/5): He describes a hybrid process where quantitative screens narrow the universe and analysts build and debate intrinsic-value estimates. The portfolio maintains 50 names, uses valuation-based sell triggers, and has long holding periods. Market valuation and style positioning (Priority: 3/5): Although the firm is mainly bottom-up, Resendis says aggregated intrinsic-value signals can inform macro views. He believes markets periodically become extreme but currently sees value and growth as much more aligned than during prior dislocations.

Key Arguments: Book-to-price and similar ratios are cheapness metrics, not true value measures, and have become detached from a coherent valuation theory. Academic factor research is weakened by looking backward with already-known information; live out-of-sample evidence is the proper standard. The investment factor is best explained by negative-spread, low-quality growers, not by a simple rule that investment lowers returns because of cost of capital. Economic profit should be judged as ROI relative to cost of capital; firms that compound above their cost of capital are the real winners. Intrinsic value should incorporate accounting adjustments, R&D capitalization, leases, inflation, and economic depreciation to better approximate economic reality. A stock’s worth depends on economic margin, growth/reinvestment, competition, and risk—not just accounting ratios. Book-to-price often works because it accidentally captures leverage, financials, and distressed/low-quality businesses, especially in small caps. Large-cap and small-cap value are not the same phenomenon; combining them can create misleading evidence and capacity issues. The firm’s five-factor model explains away many traditional anomalies while preserving significant predictive power in its own variables. Markets can be over- or under-valued for long periods, but expected returns are determined by current price versus intrinsic value, not by past underperformance.

Data Points: History calibrated for book-to-price studies: 1963–1991 - Referenced as the sample period used in the original Fama-French analysis. Post-calibration performance window: Since 1991 - Resendis argues book-to-price performed much worse once it became publicly known and live. Applied Finance valuation framework origin: 1995 - The firm says it calibrated its valuation models in 1995 using historical data. Live production start: 1998 - Resendis says the model became consistently produced in real time around 1998. Cisco intrinsic value example: $14–$15 per share - Their estimate during the tech bubble when Cisco traded around $80. Cisco market price example: $80 per share - Used to illustrate bubble-era valuation extremes. Amazon 2017 net income: $3 billion - He compares accounting earnings to economic cash flow estimates. Amazon 2017 operations-based cash flow: Almost $30 billion - Illustrates how accounting adjustments materially change economic profitability. Model coverage: 20,000 stocks per week - Approximate scale of weekly valuation work across U.S. and global equities. Total model runs: Over 20 million times - Represents the depth of the historical dataset and repeated weekly valuations. Price convergence estimate: About 15% per year - University of Michigan researchers estimated average convergence of price to intrinsic value using AFG data. Portfolio size: 50 names - The Valuation 50 strategy holds exactly 50 stocks at all times. Average holding period: About 12 years - Reported for the long-term, low-turnover portfolio. Turnover from acquisitions: About 20% - A significant share of turnover comes from mergers and acquisitions. Market-call history: 4 calls total - Three explicit market calls plus one implicit call through research output. Explicit market call periods: 2008–2009 and March 2020 - The firm identified major dislocations as buying opportunities. Recent relative-value stance: August of the current year, then retracted two weeks later - They initially said value looked cheap relative to growth, then later judged the gap much smaller.

Pivotal Quotes: "quantitative value investing has no identity" — Rafael Resendis: Summarizing his critique that traditional quant value has drifted from intrinsic worth into factor hunting. "we believe first you have to really answer how well is the firm performing before you can attempt to say what is the firm worth" — Rafael Resendis: Explaining why economic margin and performance measurement come before valuation. "the riskiest firm to own is the smallest, most highly levered firm in the stock market" — Rafael Resendis: Describing Applied Finance Group’s view of risk and leverage in valuation and cost-of-capital modeling.

Implications: For investors, the message is to distinguish true intrinsic value from cheapness screens and to test strategies on live, out-of-sample evidence. The conversation also warns that growth, intangibles, and financials can be misread if valuation ignores economic profitability and reinvestment quality.