Part 3 in a series of tutorials on value investing : How to assess the earnings of a company.
Part 1: My Investment Philosophy
Part 2: The General Store
Part 3: The Earnings Equation
Part 4: Building a Valuation Framework
Part 5: Hitting The Books
Part 6: When To Sell
The Earnings Equation
If the price : earnings ratio is my north star when it comes to valuing a potential new purchase, then the earnings part of that equation takes on enormous significance. The price part of the ratio is not up for debate. It is simply the current share price. You can get an up to the minute stock quote from the internet in seconds. The same can’t be said for the earnings part of the formula.
If you look at a list of stocks in the newspaper or pull up a stock’s summary page on one of the big financial websites, you’ll always see the p:e ratio prominently displayed. Ignore it. The p:e ratio as quoted by these services is so frequently wrong that it is virtually useless to you in your investigation of a company.
The quoted p:e ratio can be wrong for any number of reasons. The earnings figure that the p:e ratio is based on could simply be several months out of date. New results may have just come out yesterday and not be reflected in whatever database you are looking at. Or the company may have issued more shares since the last time the data was updated. More shares means lower earnings per share and therefore a higher p:e. The earnings could be technically accurate but include one-time items that distort the picture significantly. For example, they could be inflated by a big tax rebate or licensing payment. Backing out these payments will give you a better sense of what earnings will look like in the future without these “one time” items. Conversely, a company may have been doing some heavy duty restructuring of its operations and these restructuring expenses may be dragging down profitability. If we want an idea of what earnings might look like once this restructuring is finished we would back these expenses out of our income calculations or, at the very least, average them out over a number of years.
In short, the earnings number that I arrive at after applying all my adjustments can sometimes be miles away from the officially reported number. As a result, the p:e ratio that I calculate for a stock can be very different from the number that other investors might come up with. Conceptually, what I’m looking for is an earnings figure that represents the ongoing, sustainable earnings power of the company I am thinking of investing in. What matters more to me than what earnings were this year or last is what those earnings will be like in the future. That is what is going to determine the value of my investment. The earnings number I come up with is designed to reflect my best guess as to what those future earnings are going to be.
In the “Hitting the Books” article in this series, I go into the finer details of how I calculate my own earnings number, but suffice it to say that I always go straight to the source. For every new company I investigate, the first thing I do is open up a copy of that company’s last few quarterly and annual reports. I flip through to the income statements and start working my way through them line by line, looking for any items that appear to be non-recurring in nature or that may be mis-stating the true underlying earnings power of the company.
Be aware that companies will often helpfully provide you with their own version of an adjusted earnings number. But they will typically be very liberal with what they choose to exclude from their earnings calculation to arrive at that adjusted figure. Just as with the reported earnings, ignore the provided adjusted figure as well. Always calculate your own version of adjusted earnings from the ground up, using the company’s financial statements.
After I’ve spent some time cleaning up the income statements, I’ll have a much more accurate picture of what is really going on at the company in question. It is surprising sometimes how a single line item in a quarterly report from a year and a half ago can completely change the overall investment picture.
Frequently, this exercise will immediately highlight the reason the company in question is not the bargain I thought it was. For example, a chain of sporting goods stores might be reporting earnings per share (EPS) of $5.25. Let’s say the stock was trading at $50 per share, giving the company a p:e ratio of 9.5. Potentially interesting and worth taking a closer look at. However, when I look more closely at the income statements, I see that 9 months ago, the company sold off an underutilized warehouse for $20 million. Doing some math using the number of shares outstanding I see that this works out to a windfall gain of $3.00 per share. Which means that absent the ability to sell off a new warehouse every year, this company would realistically be expected to make only $2.25 per share going forward, not the $5.25 they reported. Calculating the p:e ratio again, using this new, more realistic earnings number gives me a p:e ratio of 22. Not nearly so appealing! Add in some additional color, like the fact that sales have been weak lately (which explains why they had an under-utilized warehouse in the first place) and I am quickly losing interest.
If I do not come across any items in the last few years that seriously derail my investment thesis, then I will go back and calculate my own “adjusted” earnings number going back even further. Ideally, 20 years, if the data is there. Don’t worry, once you get the hang of it, this process only takes 10 minutes or so. And after you’ve done it once, you can come back to this investment in the future without having to re-calculate everything from scratch. After finishing, you’ll have an accurate and very comprehensive picture of how this company has evolved over the years. This will let you move on to the next step of the earnings equation.
Baseline Earnings
Once I have a full set of adjusted earnings numbers calculated out, I need to determine which of those numbers to use in my p:e formula. The number I end up choosing is what I call my “baseline” earnings. Most often it will be the most recent one, the “trailing” 12 month earnings number that you get from adding up the earnings of the last four quarters, but there are cases when using another time period makes more sense. For example, during the height of covid, companies went through a period of a couple of years where they were insanely profitable. Government stimulus drove record spending at the same time that supply chain disruptions meant there was less to buy. Prices shot up and many companies capitalized on this. Earnings were high while cautious investors kept the share prices low, expecting those windfall profits to dry up. As a result there were quite a few low p:e stocks to take a look at. During this time, I often went back and used an earnings number from before the pandemic began as the baseline earnings number that I plugged into my p:e calculations. This steered me clear of a lot of potential value traps. (Situations that appear to offer good value because of a low p:e but are not really true bargains.)
Other times, the most recent trailing 12 month set of earnings will be misleadingly depressed. The most obvious example of this is during recessions. Recessions are an unfortunate fact of life and when they occur, company profits can take a nosedive. Especially if they are the smaller, more cyclically sensitive types of companies that I favor. But these downturns never last forever and if you relied on trailing earnings only, you’d miss many a good turnaround opportunity. At the depths of the recession, earnings will likely be in the toilet. The companies may even be losing money. The p:e ratio will either be very high because earnings have dropped so low or be incalculable because there are no earnings to begin with. In these cases, I’ll go back and use an earnings number from before the recession began as the basis for my p:e ratio. These pre-recessionary earnings are referred to as “peak earnings” and the p:e ratio you get by using them is called the price to peak earnings ratio or p:pe for short.
I’ve found a lot of good investments this way but you do have to be careful. Some companies never fully recover. Some even go bankrupt. To minimize your chances of going down with the ship, it pays to do your research and keep your eyes wide open. I wouldn’t recommend investing in these turnaround type plays to the beginner, but once you’ve gotten some experience under your belt, it can be a lucrative sector to play in.
Another situation in which I’ll stray from simply using the most recent set of trailing earnings is when earnings have been very lumpy over time. Some companies, often ones in commodity type industries like chemical manufacturing or oil and gas drilling, will have earnings that bounce up and down like a yo-yo. Most of the time I’ll just ignore these companies and move on to something more stable and predictable, but if the fluctuations are not too extreme, I’ll also sometimes take the average of a representative span of years and use that as my baseline earnings number.
The Exception That Proves The Rule
The key takeaway here is that the earnings number you use to calculate your p:e ratio is malleable. The number the company reports or that a Wall Street analyst uses or that you see quoted in the newspaper is not necessarily the number you want to use. You need to run your own calculations and use your own judgment.
The point is to generate a number that best represents the future, sustainable earnings power of the company you are analyzing. Often that will simply be however much money the company made in the most recent period, but not always. Those exceptions can save you a lot of grief by steering you clear of value traps or lead you to interesting opportunities that you might otherwise have missed.
In Part 4: Building a Valuation Framework, we’ll expand on the simple p:e ratio by including an estimate of future growth and thereby create a usable framework that we can use to home in on those rare, undervalued opportunities.