Earnings Predictability — Understanding Valuation

§ 01

Predictable earnings come from: (1) recurring demand, (2) pricing power, (3) low input-cost volatility, (4) stable competitive position, and (5) minimal regulatory shocks.

§ 02

Company	Earnings Volatility (10yr)	Why
Coca-Cola (KO)	Very low — EPS grew steadily every year	Same syrup, same brands, recurring global demand
Procter & Gamble (PG)	Low — dividend raised 65+ consecutive years	Consumer staples, pricing power, broad distribution
Delta Airlines (DAL)	Extreme — losses in 2020, strong in 2019 and 2023	Fuel prices, labor, pandemics, fare wars
Exploration E&P	Extreme — tracks oil price swings	Commodity exposure, capex-heavy, no pricing power

§ 03

Pull up KO and DAL side by side. Look at 10-year EPS history on each Ticker view. Notice how KO's line is nearly straight while DAL's swings wildly — that visual difference IS earnings predictability.

§ 04

Why Buffett pays more for predictable earnings: a DCF is only as reliable as its inputs. If you cannot forecast next year's earnings within 10%, you certainly cannot forecast year 10. Predictability shrinks the error bars on intrinsic value — so the same 15% discount rate produces a tighter, more actionable valuation. That confidence is worth paying for.

§ 05

Which is MORE important for valuation confidence: last year's EPS, or the standard deviation of EPS over the last 10 years?

§ 06

Practical rule: if you cannot forecast earnings within +/- 15% three years out, demand at least a 30% margin of safety on intrinsic value — or skip the investment entirely.

Five questions · AI feedback

Sit with the ideas.

In his 1996 Berkshire Hathaway Chairman's Letter, Buffett wrote that he would prefer a lumpy 15% return to a smooth 12%. Yet his largest long-term positions (KO, AXP, MCO) all share predictable earnings. How do you reconcile these?

Why:

Sit with the ideas.

Other lessons in this course.