Notes on Value Investing: From Graham to Buffett and Beyond

2025-11-19

This is an English translation of a reading summary.

Foreword

  • Graham and Dodd published Security Analysis in 1934, creating the field of value investing. The ideas were further elaborated in The Intelligent Investor in 1949.
  • In 1950, Warren Buffett decided to study under Graham and Dodd at Columbia University.
  • Another student, Mario Gabelli, convinced Roger Murray (co-author of the fifth edition of Security Analysis) to teach value investing to his analysts. The author of this book, Bruce Greenwald, later studied under Murray and then partnered with him to teach a course at Columbia University, which attracted countless students.

Chapter 1: Principles

  • Modern investment theories of the 1950s and 60s, which posit that financial markets are fully efficient, have faced numerous challenges over the years. These include clear examples of investors beating market indices and the rise of behavioral finance, which argues that investors tend to overreact.
  • Value investing is founded on three key market characteristics:
    1. The market’s mood swings, like “Mr. Market.”
    2. The frequent divergence between a company’s value and its stock price.
    3. The “margin of safety” created by this divergence, which can generate superior returns.
  • Fundamental analysis (which excludes technical analysis) falls into two broad categories: focusing on the overall economy or on specific securities.
  • Among those who focus on specific securities, most analyze the interaction between a stock’s price history and related economic variables, such as management changes, growth needs, financial leverage, M&A, asset sales, etc.
  • The majority of these investors focus on corporate earnings: buying when earnings beat expectations and selling when they don’t.
  • This differs from value investing in two ways:
    1. Their method emphasizes future price movements based on expectations, not the relationship between current price and intrinsic value. Its conclusions can be applied to stocks with very different P/E ratios.
    2. Their method does not provide an identifiable margin of safety.
  • Three pieces of evidence support the effectiveness of value investing:
    1. Mechanical Screening Tests: Systematically buying stocks with low P/E and P/B ratios has been shown to generate average annual returns 3-5% higher than the overall market since the 1920s.
    2. Institutional Performance: Large institutions that employ value investing, such as Oppenheimer and Tweedy, Browne, have significantly outperformed the market.
    3. Top Managers: The majority of fund managers who have consistently beaten the market over the long term are what Buffett calls the “Graham-and-Doddsville superinvestors.”
  • Some argue the superior returns come from higher risk, but this is incorrect. Using standard academic risk metrics like the volatility of annual returns or beta, value investing portfolios typically show lower risk than the overall market.
  • Even by other standards—such as the extent of price drops on bad news, declines during bear markets, or maximum drawdowns—value portfolios are less risky than the market.
  • To avoid getting lost in the market, value investing relies on a three-part process: a search strategy, a valuation methodology, and a portfolio strategy.

Chapter 2: The Search for Value

  • Stock-picking approaches generally fall into three categories: pure fundamentals (ROE, ROIC, EPS growth, revenue growth), technicals, and relative valuation (P/X ratios).
  • Value anomalies persist due to behavioral biases:
    • Institutions are constrained by policies, regulations, and scale (e.g., it’s difficult for them to invest in small-cap stocks).
    • Fund managers are limited by company policy and are influenced by herd mentality (“Nobody ever got fired for buying IBM”).
    • Individuals are subject to various logical fallacies and cognitive biases.

Chapter 3: Valuation

  • The Discounted Cash Flow (DCF) method, which calculates present value based on net cash flows, is deeply flawed despite its mathematical precision. The estimates for cash flow growth in the first decade, and especially the perpetual growth rate for the terminal value, are bold assumptions. The further into the future, the greater the deviation, and small initial errors lead to massive discrepancies.
  • Investors have tried to overcome this. A simpler approach is using valuation multiples, like P/E, EBIT, or EV multiples. This is equivalent to a DCF but with simplified assumptions; it’s just another way to calculate a discounted terminal value.
  • Another common approach is extensive sensitivity analysis, considering a wide range of possible values for operating parameters (revenue growth, margins, etc.). However, the resulting valuation range is often too wide, and the complex interdependencies between parameters make it hard to determine which valuation is more accurate. This approach highlights the uncertainty of valuation rather than solving it.
  • Graham and Dodd’s intrinsic value method avoids these problems and has three elements:
    1. Asset Value: For a declining business, this is the liquidating value (current assets convertible to cash within a year, minus all liabilities). For an ongoing concern, it’s the reproduction cost. Without barriers to entry, market forces will push market value close to reproduction cost.
    2. Earning Power Value (EPV): Calculated as Adjusted Current Earnings x (1 / Current Cost of Capital, R). “R” is the return required to attract capital. Adjustments are made to earnings (e.g., removing one-off charges, normalizing for the business cycle) to calculate distributable cash flow. EPV is less reliable than asset value but more reliable than DCF.
      • If EPV ≈ Asset Value, it suggests the business has no competitive advantage.
      • If EPV < Asset Value, it suggests overcapacity or mismanagement.
      • If EPV > Asset Value, the difference reflects superior efficiency or franchise value. Growth only creates value if EPV is sustainably higher than asset value.
    3. Value of Growth: The difference between EPV and the total value including the franchise’s growth potential. This is the most difficult element to estimate. In a perfectly competitive market, growth creates no value because the return on new investment just covers its cost.
  • Operational inefficiencies tend to revert to the mean. Therefore, only a true franchise can create value through growth, but this value is difficult to quantify.

Chapter 4: Asset Value

  • Liquidating Value Assessment: Accounts receivable might only be 85% collectible; inventory value depends on its nature (e.g., 50% in an example); plants and property also vary (offices are more valuable than specialized chemical plants, e.g., 45% in an example).
  • Reproduction Cost Assessment: Inventory should be adjusted for price trends; real estate needs to be marked to market; deferred taxes should be discounted to present value. The book value of a plant might be zero after 30 years of depreciation, but its reproduction cost could be higher. Goodwill on the balance sheet may be overstated. Hidden assets like the value of R&D, customer relationships, and IT systems also need to be assessed. Debt also requires evaluation.

Chapter 5: Earning Power Value (EPV)

  • Example: A company earns $10M annually. If shareholders accept a 10% return, the EPV = $10M / 10% = $100M. If its asset value is $40M, the $60M gap (franchise value) will eventually be narrowed by market forces.
  • The example of Mercedes-Benz shows that even immense brand prestige and pricing power do not guarantee superior profitability (its pre-tax ROA was only 7.2% from 1995-97).
  • Only when a business has advantages that new entrants cannot replicate—a “franchise”—can it create value.
  • Franchise sources include: government licenses, patents, copyrights, proprietary know-how, downward-sloping learning curves, unique low-cost strategies, customer habit or addiction (e.g., Coca-Cola), high switching costs (e.g., Microsoft Office), and economies of scale (e.g., Intel).

Chapter 6: Case Study: WD-40

  • WD-40 outsources almost everything; it’s a marketing organization with no secret formula or patent, yet it has exceptionally high returns on sales, assets, and equity.
  • Its adjusted ROE in 1998 was 52%.
  • Adjusted post-tax EBIT was calculated to be $22.1M.
  • Using a cost of capital (R) of 8% (government bond yield + 2% premium), the adjusted EPV = ($21.9M / 8%) + $14M (net cash) = $288M. This equates to $18.51 per share, far above its book value, indicating a significant franchise.
  • Reproduction cost was estimated at $181.2M (including capitalizing 3 years of SG&A). The difference between EPV ($274M ex-cash) and reproduction cost ($181.2M) is $94M, representing the franchise value.
  • The franchise is protected by customer switching costs and the fact that it would be unprofitable for a new competitor to enter the market (a hypothetical competitor capturing 25% market share would have a -11% ROI). DuPont, GE, 3M, and others have tried and failed to dislodge WD-40.
  • The business requires little reinvestment, so a Dividend Discount Model is suitable. In 1998, this yielded a value of $25.60 per share, 60% higher than its EPV, suggesting untapped pricing power. A price around $18-19 would offer a margin of safety.

Chapter 7: Case Study: Intel

  • For Intel, the main asset value adjustment is for Property, Plant, and Equipment (PPE) and R&D.
  • R&D can be treated as a capital investment and depreciated over 5 years.
  • Marketing expenses can be estimated based on a 5-year average percentage of revenue, assuming a new entrant would need to spend this for 3 years to compete.
  • Applying this method to calculate an adjusted price-to-book ratio successfully identified Intel’s three best buying opportunities in Jan 1982, Aug 1986, and late 1988.

Chapter 8: Diversification and Risk

  • Value investors hold fewer stocks (“fewer eggs in their basket”) because they only invest within their circle of competence.
  • The margin of safety is their primary risk-reduction mechanism.
  • Most value investors avoid shorting stocks to hedge.
  • Price volatility is not the same as risk.
  • If no suitable investments can be found for managed funds, index investing is the best default option.

Chapter 9: Warren Buffett

  • Focuses on excellent businesses: strong franchises with above-average ROE.
  • Humble about his limits, staying within his circle of competence.
  • Includes excerpts from his letters to shareholders.

Chapter 10: Mario Gabelli

  • Contributed the modern concept of “Private Market Value” (PMV): the price a savvy industrial buyer would pay for an asset, which equals intrinsic value plus a control premium.
  • Unlike passive investors, industrial buyers can enact fundamental changes.
  • His team developed analytical tools to find gaps in GAAP accounting and looks for “catalysts” to close the gap between market price and PMV.

Chapter 11: Glenn Greenberg

  • “Put all your eggs in one basket and watch the basket very carefully!”

Chapter 12: Robert Heilbrunn

  • Learned from Graham to hunt for bargains.
  • Later evolved to investing in other outstanding value investors instead of picking stocks himself.

Chapter 13: Seth Klarman

  • Author of Margin of Safety. Prioritizes risk assessment above all else.

Chapter 14: Michael Price

  • Combines “street smarts” with investment wisdom.
  • Adept at finding value opportunities in bankruptcy proceedings.

Chapter 15: Walter Schloss

  • Focuses on assets, earnings, and enterprise value.
  • A diversified investor who always maintained a margin of safety.

Chapter 16: Paul Sonkin

  • “Buy what’s broken.” Seeks out unpopular and out-of-favor stocks.

Finished reading on March 18, 2020