« Video and Pack Rock (In Theory) | Main | Google: The Bear Case »

January 07, 2006


Brad Twohig

I completely agree. Its getting ridiculous. GOOG=$2000/share, not even realistic.

I went off on it on my blog, the sametime you did.


Having read about the famous Blogett I was expecting to find a blog full of wild and amazing speculations. This is one of the most well written, simple, pointed explanations of the stock valuation game. People who write about valuations (the press) often profess astonishment at valuations. Like you say, anyone who has actually played with dicounted cash flow valuations (the way people model future value of a stock) know that they're extremely sensitive to changes in your assumptions about a company's growth. Assume google will grow only 20% for the next 5 years. It's only worth $200 or less. Assume google can grow 35% or 40% per year for the next five years, then it's worth $700 or more. The point is, we don't *know* what it's growth will be. Every quarter we get another datapoint, but that doesn't make speculation much easier. The only thing I think is sensible, is to look at the macro factors - search is in a huge secular growth phase, china is a huge potential new market, and google has a great set of engineers. Those things tell me that i should have a *position* in google, but it doesn't tell me, like you point out, whether i should buy at $400 and jump out at $683. People who think like that always end up with no money, because they're *gamblers*.
Thanks for the great post!

Stuart MacDonald

I concur with the above comment. Like most of your stuff, Henry, this is thoughtful and very well written. Parking everything you've gone through, it is great that you are back sharing your thoughts again, and so insightfully to boot.

Now, how about *you* writing the "GOOG to zero" piece? :-)

- Stuart



One issue is, going back decades, the 'media', in their effort for readers and, thus, ad revenue, is eager to 'grab' at people by the heart, the gut, lower still. In particular, they like to present their 'stories' as spectacular, exceptional, and, thus, worthy of attention.

For actual thinking, evidence, reasoning, support for conclusions, primary references, even descriptive statistics as summary descriptions of situations, even just for the usual high school term paper writing lessons, the media regards such content as boooooooooring and snips it out.

So, to get a better 'story', the media likes to 'cherry pick' the data. In any serious writing, e.g., medical research, get caught 'cherry picking' the data and can end up in a very different career.

With wildly strong contrast, in serious fields, e.g., research in mathematics, physical science, engineering, the conclusion is secondary and the real interest is in the supporting evidence. Or, for an interesting theorem, want to see a rock solid proof (in the style of N. Bourbaki) right away. Without the evidence, the conclusion isn't even useful to join the newspaper kitty litter. The media, of course, cares only about what can be made to look like spectacular conclusions and never wants to consider the evidence.

Or, as the media distributes their 'content' over the Internet instead of on paper, think of the resulting boom in sales of clay-based kitty litter!

The 'dynamics' of bubble blowing go way back, at least to the tulips, and are fairly well understood. In this effort with industrial hot air generators and railroad tank car loads of soapy WonderBubble, the media has been an eager colleague.

A more complete look at the data should be at least illuminating, possibly useful. Of course, to the extent that more illumination is harmful to the bubbles, the media will not be interested on the 'upside'. But, part of what the media likes is also the 'downside' -- first build it up; later tear it down; get two, count them, two stories in one! So, in time the media may be interested in applying their cherry picking to downside 'stories'.

Yes, much of the work in the 'rational' evaluation of stocks is of the form: There is a whole lot of data that we don't have and essentially never really can have but which if we did have would let us do an involved calculation and come up with an expected value for the stock. And if pigs had wings, then we could make money selling really strong umbrellas. Sure: If make enough assumptions about enough aspects of the situation being 'stationary', etc., then at times can argue that if collect data long enough can make estimates that converge and have useful accuracy. While such things can work in some problems, tough to do with stock prices. With the meager data we do have, can find that to cover nearly all the probability mass, have to have a range as wide as $25 to $1000 like you did.

For something that could be done and might be useful:

o Range. Have (to avoid cherry picking, in advance) a collection of stock analysts. Collect the analyzes of all of them and compare over time with the other analyzes and also, later, with reality. Then, in this context, for a particular stock, say, Google, report what ALL the analyzes in the collection say, hopefully over time. So, get the 'range' of analyzes. Maybe this would be called a 'consensus' approach.

o Business Model. For a company such as Google, look at the business model, what it's been, what it is, what it's been yielding in revenue, and what it might be. I.e., at least have some 'scenarios' with some contact with reality. Maybe this would be called a 'fundamental' approach.

o Empirical. Take revenue, earnings, and growth rates in these along with stock prices, build an empirical statistical model, and use the model to estimate Google's stock price and, of course, also the confidence interval of the estimate. Of course, that interval might be your $25 to $1000. If so, then that, too, would be interesting. Maybe this would be called a 'quantitative' approach.

And, easy enough to identify other approaches.

Should the media really want to have some possibly useful information on the price of Google, such content would be crucial.

That we can't know anything at all even reasonably solid about the price of Google strikes me as giving up too soon.

Peter Cranstone

Imagine this - all future web sites encrypted their web pages, and customers who entered a simple key (stored on their PC) could access the web page. Search would dry up overnight. I don't this scenario will happen, but then DEC laughed at the PC and see where we are now.

Google hasn't faced any serious challenges yet. You have to remember they are an advertising company based on Search. You'll see them getting serious when their search improves and they stop doing all the other distracting stuff whic doesn't generate any revenue whatsoever.

Finally remember Microsoft -it's grown a new $4 billion dollar business every year for the last 4 years. The law of large numbers will eventually catch up to Google as well.

Joe Hunkins

Great insights. It's great to see guys like you sharing the informational wealth.


When Enron launched Enron Broadband, their stock shot through the roof. The assumption (at the time) was 'whatever Enron goes into, they'll make money at it.' Google has launched a bunch of products that have yet to generate significant revenue (video, earth, base, pack, news, local, orkut, answers, etc) and many of these launches have resulted in stock jumps on the assumption that Google will someday make money off these. Granted, these are actual products - unlike Enron - but the blind euphoria is the same.


Most analysts are really salespeople. So I wouldn't listen to an "advice" from them.

Make your mind up for your self. And if your intelligent and have patience stay away from companies that you can't valuate or whoose stock prices can't be justified against the valuation, like for example Google.

Buffet and Graham would never buy Google because they could never valuate it with confidence or understand what it's doing.

Tony Ninan

GOOGLE is a good company and so is its stock. The problem here is the unreasonable bubble developed in the P/E ratio. At a PE ratio of 40, or price of $314 per share, a lot of disaster in the stock market can be saved by avoiding the bubble burst.

Look at the real estate world. We still don't learn from the bitter experiences?

The comments to this entry are closed.