Google May Finally Have Grown Into Its Britches

I happen to like Google (GOOG). I like the business, I like the growth, I like the strong and clean balance sheet.

As a result, I find myself wistfully thinking about Google every so often. So I found it interesting yesterday, when Google was on my mind, that there was a write-up in the Wall Street Journal about the stock.

The column was in the Personal Finance section of the Personal Journal, and there were a couple points from the article that I had a bit of trouble swallowing. For one, the author makes the claim that Google has "a rather modest forward price/earnings ratio of 35." Now I don't know about anybody else, but for me, a P/E of 35 is a far cry from modest, especially when you're talking about a company with a market cap pushing $150 billion.

Maybe what he meant was when compared to the forward P/E's of, say, Yahoo! (YHOO) at 56 and Amazon.com (AMZN) at 70, Google's seems more reasonable. But Google is also almost four times as large as Yahoo! and five times the size of Amazon.

The other area of the article I cringed at was his contention that Google has a natural monopoly. Without getting into a long, draw-out discussion, I just don't agree there. Google has a very strong position and a high market share in a young and fast growing industry, but I would not call Google's search business a natural monopoly. I think we are only in the first inning of the story of search and Google, and though it's been hitting the cover off the ball so far, it still has its work cut out for it to maintain market share and profitability.

However, all that said, I tend to agree with the conclusion that the author reaches: "should Google drop below $450, I'll be clicking 'buy.'"

Part of my periodic check-ups on Google include taking the temperature on valuation. Google and I had a little fling around this time last year when the whole market sold off -- the stock dropped well under $400 and I just couldn't help myself. Of course, when it ran right back up, I went ahead and took the quick gains. Not bad considering it's been basically dead money since then.

At this point I wouldn't mind the opportunity to step in on Google to hold some for a good long while. We'll say more of a 'serious relationship' than another fling. Given the growth it's seen and the amount of cash the business produces, I don't think the price right now is wholly unattractive (yes, despite the 35 P/E). The problem is that you have to be on board with an average of 30% growth over the next five years.

I think that it could be worth the risk, but there's even more need for protection when you're talking about assuming growth rates like that. That's right, I'm talking about the good ol' margin of safety. At its current price there's some buffer, but if it did happen to fall 5% to $450 (or even a bit lower), there might just be enough there to let you sleep well at night.

Of course, if you buy almost any stock low enough you'll probably sleep OK -- the question is whether Google will actually get there. And the answer to that question is one that can only be answered by time... and Mr. Market, I guess.

GOOG 1-yr chart:

Comments

[John Lazerow]

Last quarter they grew at 60% AND invested majorly - $600 million in infratsructure, 1600 new employees in the quarter alone. Not only are they growing much faster than the market is giving them credit for, but they have levers to pull, and ones they can control, should the topline slow down a bit (i.e. not invest $600 mln or hire fewer people). Any stock trading at roughly a 0.5 PEG and is in a secular growth market looks good to me!

[Brian Croner]

Keywords distributed across dozens of computers, or hundreds even, make search possible for mediocre programmers. I think when one states their infrastructure development as a sign of power, that can play out to only be half the case. Without hundreds of computers to do their search, I'm not convinced they can do much of the web. I think their algorithms suffer from computational complexity drawbacks, meaning at a certain point they have to add a half a billion in new machines to handle double the size of the customer-data workload. If I had 100 machines to handle 1 billion websites, I could just make a separate entry for each word in a single table on a database per computer. Then take each web page and take out the words, one by one, and make a whole new table for each new word. Using unsigned 32 bit integers, I could index perhaps 4 billion web pages, or if I double the infrastructure, 8 billion web pages. So my guess is their algorithm is extremely simple. Using a single table, they make a separate entry for each parsed keyword (an Adword). Using another table, they make a single entry for every url. Using a third table, they pair the integer from the keyword table with the integer from the url table and attribute the pair with a word offset. And they make money from keyword arbitrage because their algorithm is not powerful enough to differentiate otherwise. To produce a sample of a document for search results, they take your keywords, if they exist in the table (without being AND, or other overused words that overpopulate their database) and just take a look into that url_keyword table. They get the index, go back maybe 4 positions and print out those, and go ahead a few as well and print out those.

That's just my observation, and from deduction I could say something like that would make sense. It's easy to do that, and I would never consider marketing that, whether or not it is actually what they use. But it doesn't scale well, so in order to facilitate growth, they MUST pay big bucks or else they simply meet increasing customer-data workload at a linear rate.

[Brian Croner]

ps- Linear at best, or a polynomial fraction. They need logarithmic complexity to be competitive, because the amount of the world data they need to "organize" doubles faster and faster. 2010 it doubles, what, every 4 hours? Try doing that with anything linear. Logarithmic or die at that point.

[anthonyc3cil]

google can not stay quiet...
______________________...
iPod Converter
www.ipodconverter.com