Of the various metrics that people use to value stocks, the Price to Earnings (P/E) ratio is probably the most popular. People often say that a stock is cheap if it has a low P/E ratio and vice versa. Indeed, many people who purport to be value investors will often buy and sell stocks by largely depending on this metric.
However, the P/E ratio is often a grossly misleading metric, and one that people shouldn't rely on for investment decisions. In this article, I'll explain what the P/E ratio is, why it's popular, and why you shouldn't rely on it.
Why P/E Ratio Is Popular
In order to calculate the P/E ratio, you need 3 pieces of information: the stock price, last year's net income, and the number of shares outstanding. Dividing last year's net income by the number of shares outstanding will give you the earnings per share. Dividing the share price by the earnings per share will get you the P/E ratio.
The P/E ratio is popular because in theory, it's an indication of a company's "earnings yield". In fact, the P/E ratio is the inverse of the earnings yield.
For example, let's say that a company's shares are trading at $10 a share, and the company earned $1 per share last year. That pegs the P/E ratio of this company at $10/$1 = 10. On the other hand, the company's earnings ratio is $1/$10 = 10% per year. If we assume that the company will continue to earn $1 a year going forward, then investors in this company can expect to get a return of 10% per year going forward.
Since the P/E ratio is the inverse of the earnings yield, lower P/E ratios indicate higher earnings yield and vice versa. This is why in theory, stocks with lower P/E ratios should outperform stocks with higher P/E ratios. But, there's a problem with this theory.
It doesn't work.
When researchers studied past stock returns, they found that stocks with low P/E ratios didn't outperform stocks with higher P/E ratios. In fact, some researchers cite this as an argument against value investing.
However, such criticisms miss the mark. Value investing doesn't call for investing in stocks with low P/E ratios. This is because there are serious flaws with the P/E metric.
Flaw 1: Tax Reversals
Many of the flaws around the P/E metric center around accounting issues. In theory, the "earnings" of a corporation should match the actual value created by the company. For example, if a company earned $100 million dollars, that corporation should have created $100 million worth of value, either by generating cash or by increasing the value of their assets.
Unfortunately, there is often a large disconnect between "earnings" and the actual value created.
One way earnings can look distorted is through one time transactions and events. For example, one very misleading way that companies can record profits is through tax reversals. Let me explain how this works.
In many developed countries, companies are allowed to deduct past losses against current gains. For example, if a widget manufacturer lost $10 million in year 1 and made $30 million in year 2, the company can apply the $10 million loss on the $30 million gain in year 2, and pay taxes on only the remaining $20 million.
Because of this, accounting rules allow companies to record part of their loss as tax assets. The thinking is that since losses can save on future taxes, the amount that the company can save in the future is an asset. However, there's a caveat.
Companies can only record tax assets if the company expects to make enough in the future to take advantage of those losses. After all, if the company expects to keep losing money, there's no longer a tax benefit to losing money.
Whether a company will make profit or not in the future is hard to predict, and sometimes, the company will make money where it had previously expected losses. In such a situation, past losses stretching back several years can suddenly become tax assets. As a result, some companies will record massive gains in tax assets in a single year, which will flow through to earnings and inflate it significantly.
As a real example of this, you can take a look at XPEL Technologies' 2010 annual financial statements on sedar.com. XPEL's 2010 income before taxes was $220,000. But by recording a tax benefit of $200,000, XPEL practically doubled their net income for the year to $420,000.
Common sense would tell us that those earnings numbers are misleading. For one, whether the company really gained an asset or not depends on its future profitability. For another, this is a one time, non-repeatable event.
Flaw 2: Asset Sales
Tax reversals are not the only one time events that can distort earnings; asset sales can, too. For example, let's say that a widget manufacturing company bought its own headquarters for $3 million a long time ago. By accounting rules, the company would have to depreciate the value of its headquarters every year, and carry the balance on its books.
Depreciation is an expense that's supposed to reflect the deterioration of an asset's value due to wear and tear. However, depreciation often misses the mark, sometimes by wide margins.
For example, the widget manufacturer may have had to depreciate its headquarters over 30 years, which means that the value of the headquarters would go down by $100,000 every year. However, the market value of the company's headquarters may actually go up during that time. Let's say that after 20 years of owning it, the company decides to sell its headquarters for $5 million. By this time, the company would have carried the headquarters at just $1 million on its book.
To account for the difference of $4 million, the company would record a one time sales gain of $4 million, and this gain would flow through to its earnings. If the widget manufacturing business generates only $1 million in profits, the company's overall earnings would be $1 + $4 = $5 million. In other words, one time sales can significantly boost the company's earnings in a single year.
But common sense tells us that this earning number too, is misleading. One can hardly count on the widget company selling additional headquarters going forward.
Flaw 3: Depreciation
In fact, this example highlights a more common and persistent problem with earnings. Virtually every company records a significant amount of depreciation every year on all sorts of assets they own. These assets include buildings, vehicles, manufacturing plants, furniture, and more. Measuring the true extent of an asset's wear and tear is very hard to measure.
Trusting the depreciation amount is further complicated by the fact that companies can choose between different depreciation schemes. Let me explain.
One standard way to record depreciation is called the straight line method. With the straight line method, you can depreciate an asset for the same amount every year. For example, if you have a vehicle that costs $35,000, depreciating it with the straight line method over 7 years would have you record depreciation charges of $5,000 every year.
Alternatively, a company can choose an accelerated depreciation scheme, which means that companies can depreciate the asset by a higher percentage earlier in its life. For example, you might depreciate the aforementioned vehicle by $10,000 each for the first two years, and depreciate the rest by $3,000 every year. There are other depreciation schemes available.
A company will typically choose between different depreciation schemes to achieve one of two objectives: minimize taxes, and manage earnings. Because companies pay taxes on their income after deducting depreciation, companies have an incentive to inflate depreciation in order to minimize taxes. On the other hand, companies know that investors often choose to invest based on metrics like the P/E ratio, so they have an incentive to minimize depreciation charges to inflate their earnings.
While companies can manipulate earnings using depreciation, it's by no means their only tool. Other accounting tools they can use include inventory valuation, revenue recognition, pension liabilities and many more.
Summing It All Up
There's a great accounting joke that illustrates this point.
A mathematician, statistician and accountant were finalists for a position as VP in a large corporation. The hiring committee asked them all the same last question:
The mathematician was first."How much is 500 plus 500 ?" , they asked"1000" he replied without hesitation."Thank you", they dismissed him.
Next the statistician."How much is 500 plus 500?""On the average, 1000 with 95 % confidence" replied the statistician"Thank you", they dismissed him.
Next the accountant."How much is 500 plus 500?""What would you like it to be?" responded the accountant.They hired the accountant. (link to the original joke)
Earnings is like the accountant's answer to the simple additions joke. They are routinely manipulated to serve the company's interests.
Now, let's get back to the reason why we started talking about earnings in the first place. The P/E ratio is by definition, the ratio of a company's share price to its earnings per share. Because the latter number is unreliable, the P/E ratio itself is also unreliable. This is why most proper value investors look past the P/E ratio, or at least take it with a very small grain of salt.
There are other much better ratios that give us a better picture of a company's value. I'll talk about some of these ratios in future articles.