Lesson 3 of 9

Base rates and the outside view

Step 3 of 6

Worked example

A story: Apple grows revenue 20% a year for ten years. Module 10 lesson 03 did the arithmetic:

1.20^10 = 6.19
$478.8B × 6.19 ≈ $2,965B of yearly revenue by FY2036

The inside view might say: new devices, AI services, more paying users. Now the outside view.

Base rate 1: Apple's own history. Its revenue grew 5.5% a year from FY2021 to FY2026 (Module 10 lesson 03).

Base rate 2: the industry average. Damodaran's table (data as of January 2026) puts Apple in Computers/Peripherals, whose average revenue growth over the last five years was 6.71% a year. It is an industry average: it tells you what is normal, not how often a rate is reached or how rare 20% is.

Base rate 3: how often firms sustain fast growth. For frequency you need a study. Louis Chan, Jason Karceski and Josef Lakonishok (NBER Working Paper 8282, 2001) looked at all US domestic common stocks in the Compustat database, with firms selected at each year-end from 1951 to 1997, and measured how their growth played out over the following one, five and ten years. One term first: the 90th percentile is the rate that 90% of firms fell short of, so only one firm in ten grew faster. Their findings:

Measure, over ten years What they found
Median growth in income About 10% a year (inflation averaged about 4% over the period)
Large firms (top fifth by market value) 90th percentile of growth in sales and in income: close to 16% a year
Firms beating the median sales growth every year 0.9% (11 of 1,265), against 0.1% expected by chance
Firms beating the median income growth every year 0.2% (2 of 1,265), about what chance would give

Read the second row slowly: in a period when inflation averaged about 4%, only about one large firm in ten grew sales faster than about 16% a year over ten years. If anything that runs high: the study counts only firms that survived, and its growth rates include reinvested dividends. Apple at 20% would need to beat it, from a base of nearly $479B.

The authors' own summary: "analysts and investors seem to believe that many firms' earnings can consistently grow at high rates for several years. The evidence suggests instead that the number of such occurrences is not much different from what might be expected from sheer luck." They found sales growth more persistent than earnings growth.

What the outside view says. Not "Apple can't do it". It says 20% for ten years would put Apple beyond the top tenth of large firms in that study, so the story needs unusually strong evidence. A forecast near 5–7%, like Module 10's, sits inside the base rates.

Named, not linked: Michael Mauboussin and Dan Callahan's The Base Rate Book (Credit Suisse, 2016) tabulates this for the world's largest companies; the official copy couldn't be checked at source, so its figures aren't used.