The Price/Earnings ratio, or P/E ratio, is a ubiquitous measure of the relative value of a company. The lower the ratio the less you pay for the company's earnings when you buy the stock. But taken by itself the P/E ratio has very little meaning. It has two big problems which I will address:
- It does not take into account the debt that a company has on its books. Two companies, identical except for levels of debt, are clearly worth different amounts. The high debt company should have a lower P/E ratio than the low debt company, but this could lead to the illusion that the low P/E ratio is a better value.
- Earnings-per-share is an accounting number and may not reflect the real amount of cash that the company generated.
In order to solve these problems, let's first look at how I view buying a stock.
There's a Company Under There
When you purchase a stock you are buying a piece of the underlying company. So a natural question to ask is "If I had the necessary capital, would I buy this company outright?" To be able to answer this question you need to determine the total cost of buying the company and retiring all of the company's debt and debt-like obligations. I'll call this the Ownership Cost. This value is simply the current market capitalization plus the value of the net debt. Next, you need to determine how much profit you can pull out of the company each year. The unlevered free cash flow is a good choice for this value, but I prefer Owner Earnings, a term coined by Warren Buffet. This is simply the sum of the Net Income, Depreciation, and various non-cash charges. From this value you subtract the average capital expenditure and add back interest payments adjusted for taxes, since interest is tax deductible. The result is the amount of cash that you could reasonably expect to pull of the your newly purchased company each year.
A Clearer Picture
Using these two values, Ownership Cost and Owner Earnings, I'll define a ratio of the two, the Ownership Cost/Owner Earnings ratio. This ratio tells you how many years it would take for you to recoup the cost of buying a company and retiring the debt, assuming no growth. This is exactly what the P/E ratio claims to do, but this new ratio solves both of the issues that can make a P/E ratio deceptive. Let's apply this new ratio to two companies, Ford (F) and Cisco (CSCO).
Ford is a large automobile company with a 17% market share in the US. In addition, Ford has the distinction of not requiring government bailout money during the recession. Ford's has a forward P/E ratio of 6, which would suggest that the company is cheap. Let's calculate the Ownership Cost/Owner Earnings ratio to see how cheap Ford really is. Here's the calculation for Ownership Cost:
To calculate the net debt you take the total cash and investments and subtract debt and debt-like obligations, including things like pensions. For Ford, the total ownership cost is far higher than simply the market capitalization due to the large amount of debt. Now to calculate Owner Earnings:
|Stock Based Comp.||$0.17B|
|Relevant non-cash charges||$-13.64B|
|Interest adjusted for taxes||$4.43B|
In 2011 Ford claimed a $11B accounting change which needs to be subtracted from net income. Capex was averaged over the last five years, and since Ford didn't pay any taxes in 2011 due to the accounting change, I simply added all of the interest back. In the future Ford will have a non-zero tax rate, so the result of this calculation will be a little on the low side. This leaves Owner earnings of just over $10B.
Using these results I can calculate the Ownership Cost/Owner Earnings ratio, which comes out to 10.6. If I use a tax rate of 35% the ratio comes out to 12.2. Note that this is twice the P/E ratio. Of course, the piece of the puzzle which is missing is growth, and a high enough growth rate justifies a higher ratio. But regardless of growth, in the case of Ford the extremely low P/E ratio is clearly deceptive.
Cisco is the world's leading supplier of data networking equipment and software. Unlike Ford, which has a large amount of net debt, Cisco has a large amount of net cash, which effectively lowers the cost of buying the company. Cisco has a forward P/E of 8.8 and a current P/E of 12.3. Let's do the same analysis as above for Cisco.
In Cisco's case the $32B in net cash lowers the ownership cost.
|Stock Based Comp.||$1.42B|
|Relevant non-cash charges||$-0.1B|
|Interest adjusted for taxes||$0.52B|
Calculating the Ownership Cost/Owner Earnings ratio for Cisco yields a result of 6.05. This is a factor of two smaller than the current P/E ratio and suggests that Cisco is undervalued.
Using the Ownership Cost/Owner Earnings ratio gets around the fuzziness of the P/E ratio and provides a clearer picture of the relative value of a company. In the case of Ford, the company is not as cheap as the P/E ratio suggests, and in the case of Cisco the opposite is true.