When it comes to predicting stock prices, we all want convenient, easy-to-use metrics that will give us the answer to our question: Can I make money in this stock? Unfortunately, like many things in life, simple and quick short-cuts can lead us to the wrong conclusions. It's been a little over a year since I attempted to debunk one such valuation tool, the PEG Ratio, which I called "silly" (perhaps a bit too sensational!). That article, as old as it is, still gets hundreds of views a week, suggesting to me that these discussions of how to value stocks are of interest.
Today, I want to tackle another popular metric, known as "Price/Sales" (P/S). To calculate the value, and here is the simple part, all you need to do is divide the market cap (price of stock times number of shares outstanding) by the trailing sales (typically the last four quarters). The result is always a positive number. You can apply this to an individual stock or even to the whole market. The S&P 500, for instance, currently trades at 1.4X its trailing sales. This is pretty much the middle of the road for the past two decades, with a range of 0.7 (early 2009) to 2.3 (2000).
While there are many circumstances that are appropriate for using sales, like when there are no earnings, it can't be used without understanding how to use it properly. The bottom line is that you can't make money buying low P/S stocks (or shorting high P/S stocks) blindly.
What's Wrong with P/S?
There are three elements that are totally omitted from the analysis when one focuses solely on the P/S ratio:
- Growth Rates
- Balance Sheet
The first objection to using P/S in absolute terms is that it fails to incorporate profit margins. Basic investment theory tells us that companies are ultimately valued on their earnings or the cash flow to their shareholders over time, discounted back to the current value. If a company has really high sales but never earns a dime, it's likely to be a poor investment in the long run. Still, without sales, you can't have earnings, so sales are important to watch. There are several different layers between earnings (which fuel dividends) and sales, and one can compare these to sales by looking at various "profit margins":
- Gross Profit Margin = Sales less cost of goods divided by sales
- Operating Profit Margin = Sales less cost of goods and operating expenses divided by sales
- Net Income Margin = Net income (operating profits less interest expense and taxes) divided by sales
The higher the net income margin, the more valuable each dollar of sales is. Sometimes, though, a company isn't profitable, so investors will look at the margins "higher up" in the income statement. The bottom-line is that one must have an understanding of the margins in order to use P/S. I will revisit this concept with examples below.
Our second objection to P/S is that it fails to account for different growth rates. Companies that are growing rapidly deserve a higher valuation than those that are not. Stocks are valued based on the future, which, of course we don't know. We can estimate, though, and, all things equal, the future earnings for a company are higher the more growth it has relative to another one. This implies that one should pay more today for a dollar of sales from a company expected to grow faster than another company.
The final objection to using P/S is that it fails to incorporate the structure of the balance sheet. In other words, simply comparing the market cap to sales misses out on the potential presence of debt (or cash). The good news is that this shortcoming can be addressed very easily: Substitute "enterprise value" for market cap. EV is defined as market cap plus total debt less cash and investments. While EV/S will still suffer from the other two shortcomings, it will allow a better comparison between companies.
In case this point isn't clear, let's consider Company A and Company B, which are similar in all regards except for the fact that company B has $100mm debt. Both companies have a market cap of $100mm and generate sales of $50mm. In this example, both stocks have a P/S of 2.0. A blind investor might be indifferent. The more astute investor, though, would quickly realize that Company A, which has an EV/S ratio of 2.0, is substantially cheaper than Company B, which has an EV/S ratio of 4X ($100mm plus $100mm is $200mm, which is divided by sales of $50mm).
So, it should be clear that there are several problems with the P/S ratio. While one is very easy to address, the first two are not. To illustrate why one can't use P/S, I want to use some examples. We will first look at two stocks with a relatively high P/S and then two with a relative low P/S.
I said earlier that the S&P 500 has traded between 0.7X and 2.3X sales for the past twenty years. A blind use of P/S would suggest avoiding "high" ratios and buying "low" ones. Now, these examples are chosen obviously with the benefit of hindsight, but they serve to illustrate a point.
Making Money with High P/S Stocks
Our first stock is one of the best stocks in recent years despite a massive correction of late:
(click to enlarge)Had you looked at Apple (AAPL) at the end of 2008 and used P/S, you might have passed. After all, it was trading at 2.3X (middle panel). While this was down from 6X earlier that year, it was still 2.6X the P/S of the S&P 500 at the time (bottom panel). Even now, the stock is back to 2.6X.
When we look at Google (GOOG), we see a similar story. At the end of 2008, the stock traded at 4.6X. While down from where it had been, this was still a pretty big number at about 5.5X the P/S ratio of the S&P 500. I include this example because it also illustrates how a stock can perform very well, while having a dramatically higher P/S ratio than the market and see that ratio change only marginally (currently at 5.2X compared to 4.6X at the end of 2008) yet the stock can more than triple! This happened because sales grew dramatically, more than doubling from $22 billion in 2008 to $50 billion last year. The market wasn't stupid! It saw high growth and high margins (GOOG NPM is about 25%, almost 3X the S&P 500's NPM).
Sometimes the sales growth can be so stunningly high that the stock goes up even as the P/S ratio declines dramatically.
You could have been the stupidest buyer ever, paying an extraordinary 30X trailing sales for Intuitive Surgical (ISRG) at the end of 2005 (when it was roughly $125) and gotten a second shot to pay the outrageous all-time high of 33X near the end of 2007 (near 360) or you could have owned it at the much lower 10X in December (near 550). You would be down in only one of those scenarios (the last one, when it was the lowest of the three occasions). The point is that really rapid-growth companies will get very high P/S ratios and you will miss out on good investments if you rely blindly on P/S. In early 2009, it bottomed at 2.6X sales, still a massive premium to the market at the time.
Losing Money with Low P/S Stocks
Just as buying a high P/S stock isn't the kiss of death necessarily, very low P/S ratios shouldn't necessarily excite you. Let's look at two examples.
First up is the company that is first up every quarter to report earnings, Alcoa (AA). Now this is actually a company where using P/S can be useful due to the cyclical nature of its earnings. Still, though, if one were to have blindly bought this stock based on the low P/S ratio at almost any time in the past five years, it would have been a mistake. Notice how low the ratio has been: Currently 0.4X, a median of 0.5X and a low of 0.2X. Even then, in early 2009, buying the stock at its low of about $5, when it was 0.2X P/S and about 30% of the S&P 500 P/S, it wasn't a great investment, at least compared to now. The S&P 500 has more than doubled since then, while AA is up 70%. To be fair, one could have sold it early 2011, but using P/S, why? It was still below 1X and still a big discount to the market. What's important here is that the market doesn't give big valuations relative to sales for companies like AA, which has high debt, low long-term growth and low margins.
The same can be said for Hewlett-Packard (HPQ). At the end of 2008, the stock traded near 36. This was a P/S of 0.7X, about equal to the S&P 500 when the S&P 500 was at its lows by this basis for the past 20 years. Bargain? Hardly! We know how this one played out, with the stock, even after a fabulous run from its 2012 lows, still way down. The P/S metric has gotten even more attractive, especially relative to the overall market. Like AA, the market won't pay a high P/S ratio for low margins, high debt and slow growth.
How to Use P/S Effectively
I have demonstrated that P/S can't be used blindly, but the good news is that it can still be very useful in many situations. If you are going to use it, you should substitute EV/S, as this more properly accounts for differences in the capital structure. If this isn't possible, then one should be careful to compare companies with similar balance sheets.
P/S (or EV/S) is most useful in situations where the company isn't generating EPS or when the EPS isn't at "normal" levels. This can apply to young companies or perhaps cyclical ones that are experiencing depressed margins. The important step one must take is to use the metric to compare stocks with similar margins and similar projected growth rates. For instance, if one is looking at "cloud companies", one can take all of the stocks growing annual sales between 30% and 50% with gross margin above 70% and then compare the P/S ratio.
While I have suggested that P/S (or EV/S) can be useful for comparing two or more companies when the growth rates and profit margins are similar, you can also look at the individual history of a company and compare the current ratio to historical levels. Again, though, one needs to be careful to consider that the balance sheet, margin potential or expected growth rate may be different today from the past.