Microsoft Price Watch

As Microsoft’s stock hit a multi-year high of $31.45 today, I decided it was time to revisit my analysis of this stock. I first bought stock in Microsoft (MSFT) on October 10, 2005 and then I bought it again on May 1, 2006 when it once again dropped below $24.50.

I originally determined that MSFT had an intrinsic value of at least $30. I provided my calculations along with those shared by some of my readers in a previous post about Microsoft’s value. Today, I decided to run another discounted free cash flow model to determine if I need to update my intrinsic value estimate.

MSFT Free Cash FlowsI once again ran a linear regression on the past 10 years of Microsoft’s free cash flows but this time it included 2006 numbers as well as the trailing twelve month free cash flows.  Then I projected out the next ten years of free cash flows. The results are shown in the chart to the right.

I then discounted those future free cash flows using a 10% discount rate. After the tenth year, I assumed a 3% stable growth rate.  That resulted in a present value of the enterprise of $242.3 billion.  To determine the market capitalization, I added in $9 billion in cash.  Then I divided by the diluted average number of shares outstanding and determined an intrinsic value estimate of $25. Uh oh, this is significantly lower than my previous estimate of over $30.  What is going on here?

There are couple of things I noticed.  First, if you look at the graph above, the most recent year’s free cash flows and the trailing 12-month cash flows are below the trend line.  If you were to mentally draw a line, it would likely have been above this one and a bit steeper (i.e., higher growth rate). From the R-squared in the graph, you can see that only 59% of the variation in free cash flow growth is explained by the trend line.  I prefer to see at least an R-squared of 0.75 before I am comfortable with the model’s results.

I checked in with the analyst estimates for MSFT, and they forecast next years earnings to grow by 16.6% and to average 12% over the next five years.  The linear regression I ran only grows at 5% over the next five years.

I also noted that Microsoft’s business is at a major inflection point. The company has just launched several major product upgrades including Vista, Office 2007, and XBox 360. They have also launched an upgraded online presence called and their own portable music and video device, Zune.  Each of these products and service have the potential to deliver significant long-term earnings.

If I substitute a 12% growth rate (a less conservative estimate) instead of using my linear regression estimate of future growth, I calculate an intrinsic value of $36.  I am not as confident in this estimate, but it is useful to create and upper bound on my estimate of Microsoft’s intrinsic value.

Given my uncertainty regarding Microsoft’s intrinsic value, I am going to stick my original estimate of at least $30 per share.  With the market’s rich valuation currently and the potential for a serious correction building, I am going to implement a stop at $30 for Microsoft. If shares drop back down to $30 or lower, especially on bad performance news, I will likely sell my shares of Microsoft. I will be closely watching Microsoft at this point and I will be updating my intrinsic value estimates as new information becomes available.

Full Disclosure: I own shares of Microsoft.

8 thoughts on “Microsoft Price Watch

  • January 18, 2007 at 2:52 pm

    I don’t understand this. You rationally determine that MSFT’s value is $25, but you apply some optimistic math to stick with your $30 valuation. But MSFT is already over $31, so you decide to keep it while it’s selling for more than your valuation, but sell it if it gets cheaper? Is this the bizarro world version of Warren Buffett’s process?

    I think you have some emotional commitment to MSFT that is making it difficult for you to think rationally about it. Clearly a 5% annualized growth rate was probably too conservative. But even if you think MSFT is worth $36, what margin of safety do you want? Right now you have less than 20%, that’s not very cheap.

    You should consider selling MSFT right now. Half of your “upside” is pretty tenuous. Zune is a terrible product. ZBox is a money drain. Vista will likely start very slow as few people need it or want to pay the upgrade costs.

    You have a company selling approximately for what you think it is worth. Don’t search for artificial reasons to hold it.

  • January 20, 2007 at 10:01 am

    I think George’s thought process on MSFT is justified. Estimates of intrinsic value are not static, as they change over time as new information arises. MSFT has made some good uses of capital – remember they tendered at $24.75 – which created value for the remaining shareholders. So I don’t see any reason to be concerned when an estimate of intrinsic value changes.

    In addition, your reference to Zune and XBox has little impact on the valuation of MSFT. Yes, it is true they are a cash drain – and I prefer the Apple IPOD myself – but the other three major businesses more than compensate. To cite these as reasons to sell the stock doesn’t make any sense.

    Regarding the decision to sell at $30, I think you can rationalize it too. First, Mr. Market can be overly optimistic at times and overly pessimistic at times. While I am not saying this refers to MSFT, the idea of “letting your winners ride” makes some sense. Also, the point about having less than 20% margin of safety has to be taken within context. For a high-quality business, a 20 to 25% discount to intrinsic value is an excellent entry point. Do you expect to buy the Microsoft’s, McDonalds at 50% of their value? – there are too many analysts looking at these names that it is virtually impossible. Secondly, the decision to sell has to be taken into context with respect to other investment opportunities. If you have identified another stock with a greater margin of safety, then it would make sense to sell and move into the one trading at a greater discount.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.