Comparing a company’s historical EPS (earnings per share) data with the historical EPS data of the S&P 500 can help investors determine whether that company is keeping up with the market, and thus deserves its valuation.
The reason that many investors look to historical EPS in order to project future growth is that if a company has proven to be able to grow EPS at a certain rate, then it could continue to have that ability in the future.
Investors should ensure that the industry has not matured (thus slowing revenue growth into the future from the past), and also that the company has maintained its competitive advantage.
Examining historical EPS data for a company is a start, not an ending, to projecting future growth rates for valuation and should be compared in context to other pertinent factors.
EPS Growth: The Importance of Context (Growth Calculations)
For example, knowing that a company grew EPS 200%+ or 0.4% YOY, in any given year, is usually not helpful because it’s such a limited snapshot.
Company earnings (and thus EPS) fluctuate, and can swing wildly from year to year—especially for cyclical companies.
So if you compared a year where EPS was extremely low due to cyclical swings to a year where EPS was extremely high from cyclical swings—you’d get an extremely high growth rate that is very impractical to expect for the future. Any part of that two-sided equation can contribute to an inflated (or deflated) EPS growth metric, whether that’s an extremely high ending point or extremely low starting point.
This phenomenon can also rear its ugly head in long term growth calculations too.
I love websites that quote long term CAGR (such as 10Y EPS CAGR), but similarly if the starting point or ending point of the formula is abnormally high or low, the calculation for 10Y EPS CAGR will be distorted and probably won’t accurately represent the company’s true growth in EPS over the long term.
Instead, I like to take a median of 10 YOY (EPS growth) data, for example, rather than rely on a 10Y CAGR frozen in a specific snapshot in time.
EPS Growth: The Importance of Context (and Compared to the S&P 500)
The next potential pitfall when examining historical EPS data for a company is to take it out of context in relation to its history or the market.
This can particularly become problematic when looking at cyclical companies.
For example, a cyclical company will often have high growth periods during an economic boom followed by low growth (or negative growth) periods during an economic contraction. If you were to take a 3y, 5y, or even 10y snapshot of a company’s historical EPS growth data which only occurred during an economic boom (or upcycle), then you might be optimistically projecting growth which might not happen if the economy were to suddenly contract.
Similarly, you could also be too pessimistic with evaluating historical EPS if a company’s results were poor during a downcycle but you perceive the upcoming period to occur during a strong upcycle.
Finally, evaluating historical EPS growth data on a company without providing context to its relation to the EPS growth of the S&P 500 can unfairly bias the positive or negative feelings you have towards a company and its historical growth.
This can go hand-and-hand with the upcycle/ downcycle problem, and also the growth calculations problem.
If the market and economy had a period of economic contraction as a whole, then a company’s historical EPS data might make the company appear to be a slower grower than it really is.
The opposite is also true. You might be surprised with the historical EPS data (below) from the S&P 500. The presence of a short term boom or bust in EPS growth isn’t always clear and obvious until after the fact (such as the ’08-’09 recession), and even then EPS growth data from year to year is hardly quoted although it can vary wildly (as it has) outside of obvious events such as recessions or bubbles.
Historical EPS Growth Data (S&P 500)
To create this data table, I took the current S&P 500 constituents (as of October 2020), and had their historical diluted EPS data (20 years, from 2001-2020) transposed in an Excel spreadsheet.
Then, I took the YOY growth rates for each. Finally, I took the Median growth rate for every year in order to weed out extreme values, in order to find the final values for EPS growth. This first table:
- Average = 7.6%
- Median = 9.6%
You can see that the two years immediately following the “Great Recession” had superior EPS growth, which makes sense and corresponds to strong recovery growth after a “steep fall” in the depths of the recession.
I put “steep fall” in quotes because though the stock market saw huge double digit drops in those years, the diluted EPS growth only saw minimal reductions for each year for the market in the aggregate—with the median values for the S&P 500 for 2009 and 2010 just barely staying in the negative single digits.
Also interesting… leading up to 2009 and 2010 the S&P 500 constituents saw double digit EPS growth for 5 years in a row.
We’ve recently seen a period of somewhat meandering growth followed by a huge surge in 2019 (17.3%), with the strongest consecutive years following the Great Recession really just occurring in two year spurts (2011&2012; 2014&2015; 2018&2019).
Note that there is some survivorship bias in these numbers, as former S&P 500 constituents didn’t make the list and thus didn’t have their historical EPS data included in the table above.
Concluding thoughts (S&P 500)
I think seeing this historical EPS data on the S&P 500 helps to provide context when examining companies and their earnings histories. It’s very clear that 5-6% periods of growth are normal, and aren’t something to worry about as long as double digit growth years are also in play for a company.
Periods of falling EPS on a YOY basis also aren’t worrisome if swiftly recovered from, as most of the S&P 500 has indicated it’s able to do based on the historical data.
Also, the declining EPS growth numbers from the last 10 years compared to the previous 10 could indicate a softening economy/ recovery; but the data could also be skewed by the lack of former constituents muddying up growth averages (with more of these erased the farther back in time you go), as the companies removed from the S&P 500 index tend to be the worst performers in terms of growth (and create “survivorship bias”).