A Closer Look at Ratios and Value

By John Dorfman
[for educational and amusement purposes only]

(John Dorfman is president of Dorfman Investments in Boston. His opinions don't reflect those of Bloomberg News. His firm or its clients may own or trade investments discussed in this column.)

Boston, Jan. 19 (Bloomberg) -- Juniper Networks Inc. makes routers used by Internet service providers. What could be more trendy than that?

Goldman, Sachs & Co. took Juniper Networks public in June at $11.33 a share (adjusted for Friday's 3-for-1 stock split). Demand for the shares was so strong that Goldman did another sale in September at $63.33. The shares closed yesterday at $118.75.

Because Juniper Networks doesn't yet have earnings, it has no price/earnings ratio. The P/E ratio -- stock price divided by the per-share earnings, usually for the past four quarters -- is a familiar tool used by investors to see if a stock is high- priced or low-priced.

An alternate tool is the price/sales ratio, which is the stock price divided by a company's per-share revenue. Almost every company, profitable or not, has some revenue, and hence has a price/sales ratio. The P/S ratio for Juniper Networks is 1,926, the highest in the computer industry. Changing the calculation method (there are a few ways to do the math) would bring the ratio down, but it would still be unusually high.

Among all U.S. computer companies with a market value of $1 billion or more, the one with the lowest price/sales ratio is NCR Corp. Long ago, NCR was known as National Cash Register. It moved into computers in the 1970s. In 1991, AT&T Corp. acquired NCR in a hostile takeover, intending to make NCR the heart of the telephone company's future computer operations. But NCR performed poorly in captivity, and was spun off to AT&T shareholders in 1996.

Value of Revenue

NCR's price/sales ratio is 0.5. Thus, investors are willing to pay $1,926 for a dollar of Juniper Networks revenue. But they are willing to pay only 50 cents for a dollar of NCR revenue. In other words, they value a dollar of Juniper Networks revenue 3,852 times as highly as a dollar of NCR revenue.

I believe the correct analytical term for such a disparity is ``insane.''

Should there be some disparity? Sure. A company that is growing fast should almost always be accorded a higher multiple (either of revenue or earnings) than a company that is growing slowly. Analysts think Juniper Networks, based in Mountain View, California, will achieve earnings growth at a 44 percent annual pace over the next five years. They predict that NCR will grow about 19 percent a year.

That justifies some disparity in valuations, but nothing like what has actually happened in the marketplace.

Oh, by the way, Juniper Networks' revenue in its latest reported quarter was $29.6 million. For NCR it was $1.5 billion. And yet investors are valuing Juniper Networks at $17.6 billion, NCR at $3.3 billion.

Role of Profit Margin

Normally, there's a second factor that makes a price/sales ratio high or low. That's a company's profit margin. A company with a high margin brings more of its revenue down to the bottom line as profit. So a company such as Microsoft Corp., with a 50 percent operating margin in fiscal 1999, should carry a high price/sales ratio, and it does.

Based on recent history, though, you can't say that the profit-margin argument accounts for the disparity in price/sales ratios between Juniper Networks and NCR. NCR has been profitable in each of its past seven quarters. Juniper Networks has never reported a profit as a public company.

Low price/sales ratios can be helpful in spotting companies that are turnaround candidates or takeover candidates. If a company has a lot of revenue, but is earning puny profits, an acquirer may take advantage of its low stock price to make a bid for the company.

Be Careful

P/S ratios can also be helpful in finding overvalued and undervalued stocks within a given industry. You have to be careful though, in using P/S ratios across industries, since some industries -- supermarkets, for example -- inherently have low profit margins. It wouldn't be wise to buy Great Atlantic & Pacific Tea Co. and short Microsoft on the grounds that the grocery chain has a price/sales ratio of 0.1, compared with 27 for Microsoft.

For this article, I looked at price/sales discrepancies in six industries (using pretty broad industry groups). The most dramatic discrepancy I found was the one between Juniper Networks and NCR. But the other five were pretty remarkable too. Here they are:

-- In aerospace and defense, Titan Corp., fetches close to five times revenue. Lockheed Martin Corp. goes begging at 0.3 times revenue.

-- In beverages, Coca-Cola Co. commands 8.3 times revenue. Pepsi Bottling Group Inc. goes for 0.4.

-- In electronics, Sawtek Inc. rides high at 36 times revenue. Arrow Electronics Inc., a distributor, slinks low at 0.3.

Insurance, Pharmaceuticals

-- In insurance, Wesco Financial Corp. (which is also in the steel service business) enjoys a P/S ratio of 16. Loews Corp. (which is also in the cigarette business and other businesses), is stuck at a lowly 0.3.

-- In the pharmaceutical industry, United Therapeutics Corp. goes for 14,425 times revenue. Distributor Bergen Brunswig Corp. goes for less than 0.1 times revenue.

I'll tell you what. Let's just check a year from now which stocks among these 12 did best. An academic would predict that the expensive and inexpensive ones will perform the same, because today's stock prices already incorporate all known information about the companies' prospects. A momentum or growth investor would probably bet that the popular stocks, such as Coca-Cola, Juniper Networks and United Therapeutics, will be the winners.

But I'm betting on the underdogs. I'll bet that at least four of the six stocks with low price/sales ratios outperform their high-priced rivals over the coming 12 months. Big, sluggish companies can hire a new, more aggressive top executive, attract a hostile takeover offer, or even (gasp) come up with a new and better product.

I don't own any of the six laggard stocks mentioned. But if somebody stuck me with a portfolio that included NCR, Lockheed Martin, Pepsi Bottling, Arrow Electronics, Loews and Bergen Brunswig, I wouldn't feel bad. That's a hand I'd happily play.

)2000 Bloomberg L.P. All rights reserved.


I didn't think they were Allowed to say things like this in public anymore.(grin)

-- Possible Impact (posim@hotmail.com), January 19, 2000

Answers

A few still squeeze past.

-- tim phronesia (phronesia@webtv.net), January 20, 2000.