In this article I use an economic profit model to value Cisco Systems (CSCO) (Cisco, or the company). After a brief description of economic profit models, I show that Cisco is currently undervalued on a static run-rate basis, as well as under a number of different scenarios by varying amounts. However, while Cisco is undervalued, it does not need to execute before its current level of investments in R&D and marketing catch up with it.
Economic Profit Models
Before addressing the valuation of Cisco, I will give a brief description of the economic profit model used in the valuation. The overall idea of the model is to determine how effective a company is at internally generating returns on capital. The first step of the valuation process is the adjustment of a company's financial statements to fully reflect the amount of capital used in the business. This involves capitalizing certain expenses such as R&D, certain marketing expenses and operating leases.
Other adjustments include allowing for the recapitalization of purchased intangibles and the reclassification of goodwill to better reflect invested capital. Capital assets are depreciated using different schedules than those used by most companies in order to better model reinvestment needs, instead of estimating a conservative value for the assets as financial statements attempt to portray. Following a calculation of capital invested in the company, an estimate is made of the value of the company's liabilities, including the market value of equity and debt and other liabilities such as unfunded pensions, operating leases and tax liabilities.
Turning to profitability, an estimate is made based on the current after-tax run-rate gross income. This estimate is not purely cash based as several important non-cash charges are considered. After adjusting the financial statements, a corporate internal rate of return is calculated which can be related to the cost of funds, enterprise value, net invested capital and in certain limited cases a growth component to create an estimate of firm value. By far, the most important driver of value is a company's current run-rate internal rate of return.
Adjusted Accounts and Market Capitalization
click to enlarge
The table above enumerates Cisco's adjusted accounts and market capitalization assuming a $15.51 share price. A discussion of each section is included below:
Firm value. Most of this section is fairly straight forward. The big adjustments are to net tax liability and other liabilities. In the first case, it is assumed that Cisco will need to pay taxes on repatriated cash held overseas. Based on 85% of cash being held overseas and a 20% tax rate which assumes a foreign tax paid credit of 15%, Cisco would owe about $8.5 billion of taxes. There is a possibility of an amnesty and I cover that scenario below.
The second adjustment has to do with deferred revenue which is essentially a liability to customers and that represents a claim on the company of $11.7 billion and is included in other liabilities.
Invested capital. The biggest adjustment here is the capitalization of R&D and a portion of marketing expenses. R&D is straightforward and is capitalized over five years. Marketing is a little less straightforward because not all marketing spend produces benefits beyond one year. A large portion of a salesperson's time is servicing existing clients and not building new business, but that time spent building new customer relationships should be capitalized. I have chosen to capitalize about 30% of Cisco's marketing spend over five years.
A small portion of goodwill has also been reclassified to marketing and R&D, and purchased intangibles are re-capitalized using R&D and marketing dollars.
All accounts are current cost adjusted and depreciated using a straight-line method over estimated asset lives.
Gross income. Net income has been adjusted by removing interest income, which is necessary since we are removing cash from firm value. The net income estimate chosen is a essentially estimated fiscal 4Q11 annualized EPS of 0.29 cents a share which is approximately Cisco's guidance minus share-based compensation and intangibles amortization. A scenario analysis is included below to give an indication of value based on higher or lower net income figures. (Note: the discretionary adjustment is part of net income).
The other large number which sticks out is the expense capitalization add-back. This is run-rate R&D and 30% of marketing expense. These are added back to gross income because they essentially become capital expenditures for the company, the depreciation of which is handled through the mechanics of the internal rate of return calculation. Lest anyone think this add-back is a boon for the value of the company, also consider that it is matched with a new asset on the balance sheet implying a lower return on assets. More on this below.
The most important number above is the modified internal rate of return (MIRR) of 11.5%. It is a product of the basic internal rate of return (IRR) formula applied to Cisco. The modified aspect of the return is due to the mathematical property of the IRR equation which reinvests all future cash flows at the same rate; however Cisco does not reinvest all cash flows -- it pays dividends, buys back stock and builds cash. The assumed return on this portion is the cost of capital. The reinvestment rate is not designed nor implies that Cisco's returns on capital are falling; it only accounts for the fact that the company does not reinvest all its gross income in new business assets. Since acquisitions are lumpy a smoothed yearly estimate is used. My overall estimate for that portion of capital not reinvested in business assets is probably smaller than reality which benefits the company.
Cisco has a very high return relative to its cost of capital which is assumed to be 5.4% in real dollars (adding inflation that implies about 8% cost of funds). Given that high relative return, the company's enterprise value and its net invested capital, on a static basis, Cisco's firm value is about 23% under where it should be. This implies a fair value on the stock of about $19.00. (Note: In the context of this model, if Cisco's firm value increases by 1%, the stock increases by 0.92% due to the large net cash position).
If it were all so simple as reducing stock valuation to a formula, I doubt much money would be made in trading the markets. The real question is, why is Cisco trading cheap and will the discounts present in the stock persist based on future events. I should back up a minute and say that I have valued hundreds of companies using this method and in normal market conditions, a very large percentage trade close to fair value. Such a large divergence is exceptional in my view -- especially when it is an undervaluation. Here are a few valuation considerations with Cisco, positive and negative.
Declining returns on R&D and marketing spend. The biggest concern I have looking at Cisco is that R&D productivity seems to be declining. Cisco's undervaluation is largely dependent on current R&D spending generating future returns. This is the mechanics of the R&D capitalization process when spending is growing. If in two years R&D spending continued at the current run-rate and profitability did not increase the Company's MIRR would decrease to 9.8% and if the same trend existed in marketing spend as well the MIRR would further fall to 9.3%. In this event, the Company would only be about 5% undervalued. I think this is the base case scenario in the market right now -- essentially investors don't believe the increases in R&D and marketing are going to generate historically high returns.
Cisco's response is to cut costs. Cisco recently committed to taking $1 billion of run-rate costs out of the Company and it's possible most of this will come from R&D and marketing. Coming out the recession, the Company hired aggressively and it's possible they overinvested given demand and can make these cuts without impacting the core profitability of the business. However, based on my personal experience with cost cutting, I take the view that Cisco will be running uphill somewhat and while they may be able to cut $1 billion, some of that value will be transferred to customers.
Assuming they are able to capture $750 million of the cost reduction the stock would be about 32% undervalued, excluding any impact from R&D and marketing productivity. Including the full impact of increased capital invested in R&D and marketing and the cost savings assuming 2/3rds were in these categories, the stock would be about 14% undervalued.
Foreign cash. If foreign cash were repatriated at 5.25% instead of 20% instead of Cisco being undervalued by 23%, it would be 34% undervalued.
In addition to the scenarios outlined above, it is certainly possible that Cisco has just hit a rough patch and that profit will improve in future quarters above and beyond any cost cutting. I should note that growth in EPS is not sufficient because you can add EPS and destroy value by simply growing assets -- the key is improving returns on assets from current levels. If Cisco can add about 6 cents a share annually to EPS without adding assets, the share price can rally about 5% and still be at fair value. Here is a table relating value under different stock prices and net income levels.
Negative numbers imply overvaluation, while positive numbers imply undervaluation.
First, I should point out that saying a stock is 20% undervalued using this model isn't the same as saying the expected return is 20%. Stocks should increase by at least the cost of capital each year which in today's inflation environment adds about 8% per annum. Some of that will be in the form of dividends. Second, there is a small impact from growth in assets when a company can invest at higher than the market rate. I'm not going to quantify it here, but the impact for Cisco and the vast majority of large companies is small, well below the 8% per annum return as a cost of funds.
Stated another way, it is rare that growth matters, what really matters is increasing returns on capital. Third, some of the adjustments I have made may seem arbitrary and many more not thoroughly explained. To keep this article relatively short there were limits to how much I could explain each adjustment. There are a number of books on the subject that can fill in the holes. I do invest in securities using this model and these adjustments.