Valuation & *Capital Allocation*
The two halves of the same problem. Valuation answers what a business is worth — and what you should pay for it. Capital allocation answers what to do with the cash the business produces. Buffett has called allocation his easiest job. The reason is that the philosophy is straightforward; only the discipline is hard.
Price determines return — and allocation determines compounding
The 1982 letter contains one of Buffett's most direct statements on price: "For the investor, a too-high purchase price for the stock of an excellent company can undo the effects of a subsequent decade of favorable business developments." The principle is structural. A wonderful business is not a free pass on valuation. The price you pay determines the returns you earn, even when the underlying economics are excellent.
The mirror question — once you own the business, what should it do with its earnings — is what capital allocation answers. Buffett and Munger have called it their main job, the one place management decisions actually compound or destroy value at the corporate level. Bad allocation can ruin good businesses; good allocation can rescue mediocre ones. Both halves matter.
Estimating intrinsic value
Three methods, three business shapes
The 1993 letter contains Buffett's working definition: "Intrinsic value is the discounted estimate of the cash flows that can be taken out of a business during its remaining life." Three practical methods of estimating it have emerged from his decades of writing, each fitted to a different type of business.
Owner Earnings × Multiple
The most common valuation method in Buffett's framework. Take normalized owner earnings (Chapter V) and multiply by the multiple appropriate to the business quality. The challenge is choosing the right multiple, which is itself a judgment about moat strength, growth trajectory, and capital intensity.
Imagine a consumer brand with a wide moat and 4% organic growth. The financials show:
- Owner earnings 5-year average: $3.0 billion
- Most recent year (above trend): $3.4 billion
- Three years ago (depressed cycle): $2.7 billion
The 5-year average of $3.0B captures the normal earning power better than either the peak or the trough. Applying the "wide moat + moderate growth" multiple of 18×:
$3.0B × 18× = $54B intrinsic value
If the business has 1 billion shares outstanding, intrinsic value is approximately $54 per share. The current market price should then be compared against that figure to determine the margin of safety.
Simplified DCF — mental arithmetic, not spreadsheets
For businesses with a clear growth trajectory, project owner earnings year by year for the next decade, discount each year's cash flow to present value, and add a terminal value capturing cash flows beyond year 10. Buffett does this as mental arithmetic rather than spreadsheet — the precision of complex modelling overstates the accuracy of the inputs.
A software company with:
- Year 1 owner earnings: $500M
- Expected growth: 12% annually for 10 years, then 3% perpetually
- Discount rate: 8%
Approximate calculation: present value of years 1–10 ≈ $5.5B; terminal value (year 11 onwards capitalized at 3% perpetual growth, then discounted back) ≈ $13.5B; total intrinsic value ≈ $19B.
Now test sensitivity. Change the discount rate to 9%: total IV drops to roughly $15B. Change to 10%: IV drops to $12B. The 25%+ swing for a 2-percentage-point input change is the warning Buffett describes. Either require a margin of safety wider than 50%, or accept that the business is not amenable to precise valuation.
Historical earnings power — Graham's tradition
For cyclical industries, any single year's earnings can mislead badly. A peak-year earnings figure produces an unrealistic high valuation; a trough-year figure produces an unfair low one. Graham's method of using a 7–10 year average earnings figure gives a fairer picture of the business's actual through-cycle earning power.
Margin of safety — sized to uncertainty
Graham wrote in The Intelligent Investor that the three most important words in investing are margin of safety, margin of safety, margin of safety. The principle is that intrinsic value estimates are always partly wrong, and the size of the gap between intrinsic value and price is what protects you from the consequences of that wrongness.
Humility about cognitive limits
Not the same as 'cheap'
Dynamic, not static
Purchase price versus intrinsic value
The 2018 letter contains a sentence that rewards close reading: "Sometimes the price offered for a fine business is so high that even a fine business can turn into a terrible investment — if not permanently, at least painfully long." Quality alone is not enough. Price always matters, even for the best businesses.
Coca-Cola in 1998 was an extraordinary business by any measure — global brand, durable moat, decades of compounding ahead. The stock traded at roughly 50× earnings. The business continued to compound its intrinsic value at strong rates over the following 15 years.
Yet shareholders who bought in 1998 earned essentially nothing for 15 years. The market multiple compressed from 50× to a more reasonable 18× over that period. The business's compounding was real; the multiple compression entirely offset it. Even an excellent business cannot rescue a foolish purchase price for a long stretch of time.
Capital allocation
The five capital allocation paths
Every CEO chooses, every year, how to deploy each dollar of retained earnings across five paths. The choices compound over decades. The 1987 letter framed why most CEOs are unprepared: "The heads of many companies are not skilled in capital allocation. Their inadequacy is not surprising. Most bosses rise to the top because they have excelled in marketing, production, engineering, administration or, sometimes, institutional politics. Once they become CEOs, they face new responsibilities. They now must make capital allocation decisions, a critical job that they may have never tackled and that is not easily mastered."
1 · Reinvest in existing business
When to prioritize
High ROIC, moat-backed, reliable organic growth
Decision criterion
Does incremental return exceed cost of capital?
2 · Acquire businesses (full ownership)
When to prioritize
Fair price, excellent management, understandable business
Decision criterion
Fair price is "the third leg" — non-negotiable
3 · Purchase minority stakes
When to prioritize
Price below intrinsic value, no full acquisitions available
Decision criterion
Berkshire's unique 'both/and' advantage over private equity
4 · Repurchase stock
When to prioritize
Share price clearly below intrinsic value
Decision criterion
Below estimated IV — never to offset dilution alone
5 · Pay dividends
When to prioritize
None of the above offer better returns
Decision criterion
Honest acknowledgement that no better use exists
Holding cash
Sometimes none of the five paths offers an attractive deployment. Holding cash and waiting for opportunities is itself a capital allocation decision — and one that requires resisting institutional pressure to "deploy." Berkshire has accumulated over $150 billion in cash during periods when Buffett saw no attractive opportunities.
The One Dollar Test
The 1984 letter introduced the cleanest single rule for judging whether retained earnings are creating or destroying value. The question is binary: for every dollar a company retains rather than distributes, does shareholder market value rise by at least one dollar?
Buffett believes the One Dollar Test is being passed at Berkshire — every retained dollar produces more than a dollar of market value over time. The 2016 letter noted that Berkshire ranked first in the United States in retained earnings in both 2015 and 2016, "by a margin of several billion dollars over the runner-up. Those re-invested funds must earn their keep."
He has stated openly that once the test stops being passed, dividends will begin. The standard is not "Berkshire never pays dividends" but "Berkshire pays dividends when retention destroys value." The former is dogma; the latter is capital allocation discipline.
Share buybacks — when value-creating, when destructive
The 2011 letter contains the cleanest statement of buyback principles in Buffett's writing: "Saying that you plan to repurchase shares 'to offset dilution from option grants' or because management has 'faith in the company' is not sufficient reason. Unless the repurchase price is below intrinsic value, continuing shareholders are harmed. The first law of capital allocation — whether the money is slated for acquisitions or share repurchases — is that what is smart at one price is stupid at another."
Value-creating buyback
- Stock trades clearly below estimated intrinsic value
- Volume increases as price falls (correct discipline)
- Decision based on explicit IV estimate, not "stock is cheap"
- Adequate liquidity preserved post-buyback
- Continuing shareholders' per-share value increases
Value-destroying buyback
- Stock trades at or above intrinsic value
- Volume increases as price rises (EPS engineering)
- Justified by "offsetting dilution" or "confidence"
- Funded by debt that strains the balance sheet
- Continuing shareholders subsidize selling shareholders
Between 1976 and 1980, Berkshire invested $47 million to acquire approximately one-third of GEICO. Over the following decade and a half, GEICO continuously bought back its own shares on a large scale. Berkshire never deployed additional capital. Yet by 1995, Berkshire's ownership percentage had grown to roughly 50% — purely through GEICO's buybacks shrinking the share count.
In 1996, Berkshire acquired the remaining 50% for $2.3 billion. The original $47 million stake had become a roughly 50× position — almost entirely through investee buybacks at sensible prices. The compounding occurred without any further capital outlay. This is what well-disciplined buybacks produce when they happen below intrinsic value over many years.
Dividends — when there's nothing better
Dividends represent the residual decision in capital allocation. When the company cannot reinvest at returns above shareholders' opportunity cost, when no acquisitions meet the criteria, when the stock price is above intrinsic value (so buybacks are inappropriate), the remaining option is distribution. The honest framing: dividends are a confession that no better internal use exists for the cash.
Buffett's 2022 letter on the Coca-Cola dividend:
"In 1994, we received $75 million in cash dividends from Coca-Cola. In 2022, Coca-Cola paid us $704 million. Growth occurred every year, just as certain as birthdays. All Charlie and I were required to do was cash Coca-Cola's quarterly dividend checks."
The dividend grew nearly tenfold over 28 years on a position that was never added to. The yield on Berkshire's original cost basis is now extraordinary. This is what compounding looks like in practice when the underlying business compounds intrinsic value and the dividend tracks it.
Tax efficiency — the Rip Van Winkle insight
The 1989 letter contains one of the most striking mathematical demonstrations of tax timing in personal finance. The setup is simple. $1 doubles every year for 20 years. In one scenario, you sell each year, pay capital gains tax, and reinvest. In the other, you hold continuously and only sell at year 20.
Special instruments
Arbitrage — independent of the market
Arbitrage was one of three investment pillars during the partnership era (1957–1969). The discipline involves profiting from publicly announced corporate events — mergers, reorganizations, liquidations — where the gap between the current market price and the announced deal price represents a return for the investor willing to bear the deal-completion risk.
What arbitrage gives you
What it requires
Bonds and convertibles — niche, but important
Buffett's basic stance: over the long run, bonds are inferior to equities. But in specific situations — particularly during crises when terms are extraordinary — they offer compelling risk-adjusted returns.
Bonds: four core principles
- Duration matters. Long-duration bonds swing more with rate moves than short-duration bonds.
- Inflation is the silent killer. Bond coupons are quietly eroded by inflation; long-duration holders bear this risk often without recognizing it.
- Crisis is opportunity. 2002 junk bond purchases (~$8B) and 2008 distressed credit (~$14.5B into Goldman Sachs and GE) produced extraordinary risk-adjusted returns.
- Beware callable provisions. Bonds where the issuer can call early are structurally disadvantageous to holders — called away in good times, stuck in bad times.
Convertibles: a hybrid tool
Convertible preferred stock combines fixed-income downside protection (the bond floor) with equity upside (the conversion right). Buffett used this structure heavily during crises — 1987–1991 deployed ~$2 billion across Gillette, Coca-Cola, US Air, and others; 2008 deployed $14.5 billion into Goldman Sachs and General Electric on extraordinary terms.
Two principles. First, the security must be attractive purely as a fixed-income instrument — the conversion right is a bonus, not the basis. Second, only transact with management you trust deeply.
In 1989, Berkshire took a convertible preferred position in Gillette rather than buying common stock. The conversion right was eventually exercised, and the investment was profitable. Years later, Buffett acknowledged that he could have purchased 60 million shares of Gillette common stock outright for the same capital and made meaningfully more money. The convertible structure had felt clever; the simpler ownership of common stock would have produced better returns.
"I could have bought 60 million shares of Gillette common... but I thought I was being so clever."
The lesson Buffett himself drew: do not let a more elaborate structure obscure the fact that a simpler one would have been better. Cleverness is not a substitute for clarity. When ordinary common stock is the right answer, take it — even if the situation seems to invite something fancier.
Applying valuation and allocation analysis on Dividend Line
The framework in this chapter becomes operational when applied to specific companies. Six features connect directly to the chapter's structure:
Multi-method valuation panel
Margin of safety indicator
Buyback discipline analysis
Capital allocation tracker
Tax-aware portfolio simulation
Historical multiple compression alerts
Frequently asked questions
Match the method to the business shape. The Owner Earnings multiple works best for mature, predictable businesses with stable margins and modest growth — pricing power is the dominant variable, not growth trajectory. Buffett uses this for most Berkshire holdings. The simplified DCF makes sense for businesses with a clear, identifiable growth trajectory over 5–10 years before reverting to a stable terminal phase. The historical earnings power method is the right tool for cyclical businesses, where any single year is misleading and a 7–10 year average gives a fairer picture.
For most public companies you analyze, the multiple method is the workhorse. The DCF is for high-growth situations where multiple compression is the central question. The historical method is for banks, miners, energy producers, and other cyclicals where current-year earnings are not representative of the cycle.