CAGR: Compound Annual Growth Rate for Sales and Earnings

Add up five years of growth rates and divide by five. Whenever the yearly rates vary, that overstates how fast the company grew. A two-year example shows why. Revenue of $100 million that rises 50% and then falls 50% ends at $75 million: up to $150 million, then half of that. The average of the two rates is zero. The business shrank by a quarter.

Growth compounds, so each year’s rate applies to a different base. The compound annual growth rate, or CAGR, answers a cleaner question: what single yearly rate, applied every year, would carry the starting figure to the ending one?

The formula

CAGR = (ending value / starting value)^(1 / years) - 1

“Years” counts the steps between figures, so it comes out one short of however many annual numbers sit in front of you. Year 0 through year 5 is six numbers. It is five years of growth. Counting six is an easy slip, and it drags the rate down.

Any spreadsheet does the exponent for you: type the ratio, raise it to the power of one divided by the number of years, and subtract one, and on a phone calculator the key for x to the power of y does the same job if you enter 0.2 for a five-year span.

What CAGR hides

The rate says nothing about the road. Take two hypothetical companies. Both start at $200 million and both end at $322 million five years later.

Year	Company A	Company B
0	$200 million	$200 million
1	$220 million	$200 million
2	$242 million	$200 million
3	$266 million	$200 million
4	$293 million	$200 million
5	$322 million	$322 million

Both have a CAGR of about 10%. Company A grew steadily. Company B went nowhere for four years and then jumped 61% in one, which could mean a big contract, an acquisition of the kind the previous lesson on revenue growth taught you to strip out, or a one-time boom that may not repeat.

You’d pay very differently for those two. So after you compute a CAGR, look at the individual years. Five year-over-year rates take a minute. They show at once whether the average describes anything real.

The start year does a lot of work

Now suppose the year you start from was a bad one. Company A’s revenue dipped to $160 million in a recession year, then recovered to the same $322 million five years later.

Same company, same ending revenue, five extra points of growth a year. Nothing about the business changed. Only the starting point did, and a starting point at the bottom of a downturn builds the recovery into the rate, so anyone quoting a long-run growth figure, including the company in its own investor materials, has a choice of endpoints that can make the number look better or worse than a fair reading of the business would suggest.

Check which years a quoted CAGR spans. Then run it again from a year or two earlier and later. If the rate swings a lot, trust it less. A span that takes in a full business cycle, a downturn and the recovery after it, gives the fairest reading, and where a young company has only a few years of filings on EDGAR, write that down beside the rate and treat it as provisional until more years arrive.

Earnings per share compound too

Everything above works on earnings per share. Say a hypothetical company’s EPS went from $1.50 to $3.00 over five years. That’s a doubling. The fifth root of 2 is about 1.149. So EPS compounded at just under 15% a year.

Put that next to revenue. If revenue compounded at 10% while EPS compounded near 15%, something besides sales pushed earnings per share up faster: wider profit margins, a shrinking share count from buybacks, or both. The reverse gap is the warning sign, where revenue races ahead and earnings per share lag, because it usually means the growth is costing more than it brings in. The lesson on price against earnings covers where EPS comes from if you need a refresher.

The same arithmetic runs a dividend growth rate. There it measures how fast a payout has risen.

When revenue grows faster than earnings per share, the difference went somewhere. The next lesson, on growth that costs more than it earns, follows it into share issuance, thinner margins and cash burn.