Skip to main content
Message Font: Serif | Sans-Serif
 
No. of Recommendations: 12
Note: Because putting tables in a post is kind of a pain, the three tables I reference are in another post, labeled CAP analysis MSFT 2. Thanks to everyone who told me how to format posts for the boards. So, view the next post when reading this. And now…

Moore, Johnson, and Kippola discuss valuation on only a basic level. There are other books and resources that delve into the topic and provide added insight that can be used in conjunction with the Gorilla Game. However, the authors discuss the notion on Competitive Advantage Period (CAP) in Chapter 4 of the book. They discuss CAP from a qualitative basis, which is obviously quite important. Anyone wanting to learn a great deal about CAP should read this chapter. Then, you should read the chapter again.

Basically, CAP is a measure of duration, or the amount of time the company is expected to earn returns in excess of its cost of capital. CAP (depending on whom you think actually created the concept) was developed before a wonderful surge in the market (for bulls) beginning in 1995, especially in the technology sector. While the concepts of CAPs still seem valid, some companies don't appear to have the underlying fundamentals to support such a lengthy CAP. Or, it could just mean that in an environment where the business world is a complex ecosystem of interrelationships, the companies at the top of the food chain command unthinkable valuations (unthinkable CAPs). The latter reason is probably why many traditional “value” investors missed out while “value” investors of a Gorilla Game nature might have made such explosive returns.

The authors do not calculate CAP, but I am going to show my understanding of the way that Dr. Alfred Rappaport calculates value duration (his from of CAP). In the Michael Mauboussin and Paul Johnson research report that introduced CAP in 1997, they referenced Rappaport who had some work going back to 1986 on this topic (if only you knew about CAP then!).

CAPs are determined by several key value drives including sales, operating margins, tax rates, fixed capital investment, working capital investment, and the cost of capital. In addition, barriers to entry, switching costs (also a form of barriers to entry), increasing returns, and the pace of technological change impact CAP as well. These issues are discussed at greater length in the gorilla Game.

It is also important to note that CAP utilizes forecasts. Most people don't like forecasts. Forecasts are always wrong. But, I don't really think that matters. What matters is what the market thinks, and if you have a differing view from the market and you are correct, then you may make excess returns.

Also, most importantly, I want to thank FoolishFloyd for working with me offline. Many times, he was way ahead of me on this stuff, and he is truly a smart person. It took me a little extra time to read the stuff I needed to read and sit and ponder what all this might mean. So, thanks Floyd.

Microsoft's CAP

CAPs use the concept of economic profit or Net Operating Profit After Tax (NOPAT). NOPAT is a pain in the butt to calculate and I have seen many ways to calculate it. So, I cheated a little bit and used MSFT's net income instead. I have seen approximations to NOPAT that may serve folks as being good enough such as Net Income + (1-tax rate)*Interest Expense. But, since MSFT has no debt, I just used Net Income.

The other key value drivers are laid out here (see Table 1 in next post):

Fixed capital investment is capital expenditures minus depreciation, and the incremental change divides that result by the change in sales. As you can see, MSFT has negative fixed capital investment, and this is likely sustainable, b/c they don't have to invest in a lot of new fixed investments when developing another piece of software.

Working capital investment is accounts receivable + inventory – accounts payable –accruals. MSFT has no inventory, and many new economy type companies have negative working capital investment, which would increase cash flow.

MSFT has $17+ billion in short-term investments (BTW, all of this data is from the end of the last fiscal year), and no debt. I used a cost of capital of 12%, which is the rate that Mauboussin at Credit Suisse First Boston has used to discount MSFT in the past (to keep it simple for me).

Then, you just input the forecasts using Value Line as an unbiased provider of forecasts (hey Value Line is finally useful for something). However, I held the investments at the current rates and I am not sure how anyone could use value line to forecast the working capital investment (although fixed capital investment is possible).

So, forecasts may look like this (see Table 2 in next post):

Once you have the forecasts laid out, then using a spreadsheet; you can just drag it across until the row that says “Shareholder Value” equals the current market capitalization of the stock. Then, you just look up to the top of the spreadsheet and it will tell you how many years into the future that is. Then, Viola! You have a CAP. I included the first few years, and then year 24, the CAP. You would just drag across the spreadsheet and it fills in automatically. But it makes no sense to post 25 columns or more of data here.

(View Table 3 in the next post)

The discount factor is (1+cost of capital)^-year (e.g. year in the first year is -1, second year -2, 24th years –24, etc).

The PV of the residual is NOPAT (net income here) divided by the cost of capital. Then, the corporate value is the NPV of the residual + Cash flow + marketable securities. Since MSFT has no debt then Shareholder value stays the same (otherwise you would subtract this from the previous equation).

Then, you just input the forecasts using Value Line as an unbiased provider of forecasts (hey Value Line is finally useful for something). However, I held the investments at the current rates and I am not sure how anyone could use value line to forecast the working capital investment (although fixed capital investment is possible).

My impression is that CAP is a great thinking tool, and I will keep refining my use of it. It is not any silver bullet, but it does help to see how these value drivers interact to create value.

Note: This is also rough b/c it is easier for me to do it this way in this space. We all have demands (or so it seems) on the Internet, but this should provide enough of a guide to try this yourself too.

Best,

TMFFuz
Print the post Back To Top
No. of Recommendations: 0
John: Thanks for your two posts on MSFT CAP. They were very infomative and helped me clear up some of the misconceptions I had from my previous reviews, including some of Mr. Buffett's two step forecast model work.

I am still confused with three items in your posts. I admit this confusion could stem from my engineering background v.s. a financial background.

1. You said I have seen approximations to NOPAT that may serve folks as being good enough such as Net Income + (1-tax rate)*Interest Expense. This one baffles me. First; if the Interest Expense is 0, as in the case of MSFT, and the symbol * is meant as a multiply function; the second term become "0" and drops out of the equation - fine so far, except MSFT still has to pay taxes on income. But if the tax rate were 100%, we would end up with the second term reducing to a value of "0" and again dropping out of the equation - Not so fine, since a tax rate of 100% would reduce any amount of income to 0. Help!

2. You said The discount factor is (1+cost of capital)^-year (e.g. year in the first year is -1, second year -2, 24th years –24, etc). I feel more comfortable with (1-cost of capital)^year (e.g. year in the first year is 1, second year 2, 24th year 24, etc.) Are these the same??? To an engineer, -2 as an exponential power is the same as saying to the 1/2 power.


3. It would appear if something unrelated to the business, e.g., Marketable securities and investments, were to increase in value over time that CAP would be shorten. Or, even if adjusted net income increased faster than forecast, with everything else remaining the same, that CAP would be shortened. Am I missing something here?

I look forward to your future refinments to help us better understand CAP, and thanks for any help to clear up my above confusion.

Best
Harold

Print the post Back To Top
No. of Recommendations: 2
Hi Harold,

1. The interest expense* (1-tax rate) is used to add back the tax shield from debt, not the total taxes paid by the firm. Therefore, since MSFT has no debt, then only net income is used, and taxes are already taken into account in net income. Hopefully that clears that up a little bit.

2. This is the way the author does it, and I am not engineer, but it does indeed work. So, I am going to attach a MSFT Excel file with the CAP spreadsheet so you can see and play around with it.

3. I have seen CAPs done several times, and the analysts always hold them constant, so I do too. I think it is hard to forecast and most people are more comfortable modeling the changes after they occur.

If income rises faster than forecasted, CAP may not necessarily shorten because you forgot about the corresponding change in market capitalization. Maybe the CAP will lengthen, maybe it will shorten. It will depend on how the market assesses performance relative to competitors by awarding different market caps.

So, the spreadsheet is attached. Feel free to ask questions.

Best,

John
Print the post Back To Top
No. of Recommendations: 0
Hi John

I understand points 1 and 3.

For some reason I am having problems downloading the Excel file. Maybe because I just upgraded my Office suite today to 2000 Professional. Will figure it out.

Thanks for the explanations.

Harold
Print the post Back To Top