Legg Mason Capital Management

January 14, 2005

M&M on Valuation
For the great enemy of truth is very often not the liedeliberate, contrived and dishonestbut the mythpersistent, persuasive, and unrealistic. Too often we hold fast to the clichs of our forebears. We subject all facts to a prefabricated set of interpretations. We enjoy the comfort of opinion without the discomfort of thought. John F. Kennedy Yale Commencement Address, 1962 1

Michael J. Mauboussin

Cash Flow Return On Investment

[email protected]

Lowest

Highest

Expected Earnings Growth

Lowest

56.9X 35.1 26.1

22.3X 20.0 19.0 19.4

18.7X 19.0 19.4 20.3

18.6X 20.0 22.4 22.3

Highest

23.5

Source: CSFB HOLT, LMCM analysis.

Price-earnings (P/E) multiples are by far the most popular valuation metric on Wall Street. Yet most investors dont clearly understand what a priceearnings multiple implies about a companys future financial performance, or how a companys price-earnings multiple will likely change over time. We offer an analytical bridge between valuation multiples and sound economic reasoning. The role of growth in valuation is completely contingent on a firms return on capital. The key for an investor is to anticipate expectation revisions.

Legg Mason Capital Management

Introduction
Price-earnings (P/E) multiples are by far the most popular valuation metric on Wall Street. A recent academic study found that 99.1 percent of analyst reports mention some sort of earnings multiple, and less than 13 percent provide any variation of a discounted cash flow model. 2 Still, most investors dont have a clear sense of what a price-earnings multiple implies about a companys future financial performance, or how a companys price-earnings multiple will likely change over time. The widespread use of relative multiples often compounds the problem. Investors frequently justify their valuation conclusions with apples-to-oranges comparisons of businesses with very different economics. Multiples are ubiquitous but remain, on balance, poorly understood. In this short piece, we offer an analytical bridge between earnings multiples (and multiples of any kind) and sound economic reasoning. We then discuss the role of relative multiples and finish with a brief discussion of the importance of price-implied expectations in valuation. One of the investment industrys all-time great thinkers, Marty Leibowitz, covered many of these topics in a series of papers and monographs over the past 20 years (much of this work was in collaboration with Stanley Kogelman). These articles have been compiled for Leibowitzs new book, Franchise Value: A Modern Approach to Security Analysis (New York: John Wiley & Sons, 2004), which we recommend highly for readers who want to delve into this valuation topic in more detail. M&M: Melt it into Your Head One sound and intuitive place to start to understand price-earnings multiples is Merton H. Miller and Franco Modiglianis seminal 1961 paper, Dividend Policy, Growth, and the Valuation of Shares. In the section that demonstrates the theoretical equivalence of various valuation approaches, they show that an investor can express the value of a company in two parts: 3 Value = steady state value + future value creation We can further define each of the terms in the right-hand side of the equation: Steady state value = Net operating profit after tax (normalized) + cash and non operating assets Cost of capital And: Future value = Investment * (Return on capital cost of capital) * Competitive advantage period Cost of capital * (1 + cost of capital) According to Miller and Modigliani (M&M), this formula has a number of revealing features and deserves to be more widely used in discussions of valuation. What are these features, and how can they inform our valuation discussion? Here are some of the applications: The formula allows you to disaggregate a price-earnings multiple into a commodity component and a franchise component. Return on invested capitals central significance in valuation becomes immediately clear. The formula reveals the magnifying impact of growthfor better or worse. The equation illustrates the proper precautions in using relative valuation metrics. The formula offers a very handy guide for assessing the expectations in a stock.

Lets take a look at these features in some more detail.

Page 2

Legg Mason Capital Management

Disrobing the Price-earnings Multiple Part I: The Steady State Value The steady state value equals a businesss worth if it doesnt create additional value and maintains its normalized earnings. 4 Note that the steady state says nothing about corporate growth, it only assumes the incremental investments will earn the cost of capital. Companies can and do grow earnings per share without creating shareholder value. This term allows us to calculate a steady-state price-earnings multiple: 5 Steady state price-earnings multiple = 1 . Cost of capital

If we assume an 8 percent cost of capitala 4 percent risk-free rate plus a 4 percent equity risk premiumthe steady state price-earnings multiple is 12.5 times. The market expects any company with a P/E multiple above 12.5 to create future value. If a company has a P/E multiple at or below 12.5, the market either assumes no value creation or anticipates that future value creation will not offset a decline in the current base business. We can also think of the steady state P/E as a commodity multiple, the level all companies reach at the end of their life cycle. Life cycle theory suggests that companies earning a return above the cost of capital will attract competition, eventually driving industry returns to the cost of capital. Over time, economic forces push multiples down to a commodity, or steady state, level. 6 Exhibit 1 shows how the warranted P/E for a company changes over time. In this theoretical example, the companys year 1 return on invested capital is 56% and the earnings growth rate is 25%. We then fade both the returns (from 56% to an 8% assumed cost of capital) and growth (from 25% to 5%) over 25 years. Note the warranted P/E is close to 70 for high growth, high return businesses. Exhibit 1: Multiple Reversion to the Mean
P/E Reversion to a Commodity Multiple
80.0 70.0 60.0 50.0 P/E 40.0 30.0 20.0 10.0 1 3 5 7 9 11 13 15 17 19 21 23 25

Years

Source: LMCM.

Page 3

Legg Mason Capital Management

Of course, we cant assume all businesses will maintain their normalized earnings. For instance, we know fewer people in the U.S. are smoking and consumers will continue to migrate from traditional to digital photography. We can adjust the steady-state model for a declining business using a variation of the Gordon growth model. A negative growth rate lowers the value of the numerator and increases the rate at which we capitalize the cash flow: 7 Modified steady state value = Net operating profit after tax (1 + growth) Cost of capital - growth The last, obvious point here is that companies trading near a commodity multiple with a stable base business and reasonable prospects for future value creation are good candidates for purchase. Disrobing the Price-earnings Multiple Part II: Value Growth Opportunities At the time of this writing, the S&P 500s P/E is about 17.0 based on consensus 2005 estimates. If the formula applies, 25 percent or more of the markets value comes from future value creation. What are the key issues in thinking about future value creation? An examination of the equations second term, which Leibowitz calls the franchise value, reveals three key drivers: The spread between the return on capital and the cost of capital The magnitude of investment How long a company can deploy capital at positive spreads

When investors and corporate managers think about valuation, they often start with growth. But the first of these drivers shows why this emphasis is totally misplaced. If a company earns exactly the cost of capital on incremental investment, the second term of the equation collapses to zero and the appropriate valuation is the commodity P/E. If a company earns excess returns, the second term is positive and the company will create shareholder value. Finally, if a company earns below the cost of capital, the second term detracts from value, and the company will trade below the commodity P/E. The role of growth in valuation completely depends on a firms economic returns. Importantly, companies can grow earnings per share without generating returns in excess of the cost of capital. 8 Proper valuation considers the return spread first, and the magnitude of the investment, or growth, second. (See Appendix B for a discussion of current trends.) Miller and Modigliani make this point clearly in their paper: The essence of growth, in short, is not expansion, but the existence of opportunities to invest significant quantities of funds at higher than normal rates of return. Exhibit 2 illustrates this point. The columns of the table are different return on invested capital levels, and the rows are varying growth rates. The body of the table contains the P/Es that fall out of the return and growth relationships. The table shows three fundamental principles. First, a company earning its cost of capital will trade at a commodity P/E multiple, irrespective of its growth. Second, companies earning excess returns trade at premium P/Es, and growth strongly amplifies the benefits of high returns. Finally, companies earning below the cost of capital trade at low multiples.

Page 4

Legg Mason Capital Management

Exhibit 2: Return on Invested Capital

Return on Invested Capital

4% 4%
Earnings Growth

8% 12.5x 12.5 12.5 12.5

16% 15.7x 18.1 21.3 25.5

24% 16.7x 20.0 24.2 29.9

6.1x 1.3 NM NM

6% 8% 10%

Assume all equity financed; 8% Cost of capital; 20-Year forecast period Source: LMCM.

Return on invested capital (as we define it here) also represents the maximum sustainable growth rate for a business excluding external financing. 9 Typically, high-return investment opportunities are scarce, not capital. The idea of sustainable growth shows why a capital-efficient business deserves a higher valuation multiple than a lower-capital efficiency business at a given growth rate. While clearly sound in theory, Exhibit 2 also applies in reality. In Exhibit 3, we took over 2,600 U.S. companies and ranked them on CFROI by quartile (columns). We then cross-ranked the CFROI quartiles by expected earnings growth rates (rows). Each resulting multiple is the median P/E of the companies that fall into that cell. For example, the companies in the highest CFROI quartile but lowest EPS growth quartile (upper right hand corner) have a median P/E of 18.6, while the highest growth companies in the same CFROI quartile (bottom right hand corner) have a P/E of 22.3. As theory suggests, higher growth adds value for high return businesses and detracts value for low return businesses. Further, multiples rise for a given growth rate as returns rise: the three columns on the right represent businesses earnings above the cost of capital. 10 Exhibit 3: Returns, Growth, and P/E Multiples: Empirical Results

Cash Flow Return On Investment

Lowest Highest

Expected Earnings Growth

Lowest

56.9X 35.1 26.1

22.3X 20.0 19.0 19.4

18.7X 19.0 19.4 20.3

18.6X 20.0 22.4 22.3

Highest

23.5

Source: CSFB HOLT, LMCM. CFROI is a registered trademark in the United States of Credit Suisse First Boston or its subsidiaries or affiliates.

Page 5

Legg Mason Capital Management

The textbook definition of return on invested capital divides an income-statement-derived earnings numerator (typically net operating profit after tax) by a balance-sheet-based denominator. An investment is a current outlay today with the expectation of a future financial return. Historically, most of the investments companies made, like capital equipment and inventory, went on the balance sheet. However, as we shift more towards a service and knowledge economy, many of the investments companies make, like R&D or training, show up on the income statement. To assess future returns accurately, investors must consider all investments, irrespective of where they show up in the financial statements. The final driver of franchise value is the sustainability of excess-return investment opportunities. As we noted earlier, economic forces drive the return on invested capital down to the cost of capital over time. This return reversion is well documented empirically. 11 A central task for an investor is to assess how long a company earning above its cost of capital can continue to find productive investment opportunities. Stock prices often reflect ten to twenty years of value-creating cash flows. Competitive strategy analysis is particularly useful in this effort. Relative and Comparable Multiple Valuation A Barrons columnist recently suggested that a Wall Street analyst had won his devoted admiration with the simple proclamation, We think Amazon.com is a retailer. Why was this seemingly innocuous statement so pleasing? Because, I have been patiently waiting for the day when Wall Street ends this nonsense of giving blue-sky premiums and special treatment to Internet-era darlings, including, but not limited to, Amazon. 12 The implication of this thinking is that since Amazon sells stuff, investors should value it like other companies that sell stuff. This simplistic thinking, while pervasive, is flat wrong. As Exhibit 2 demonstrates, company valuation should not be based on what the company does, but rather what its economics look like. With rare exception, valuation disparities between companies within the same industry reflect legitimate differences in expected return on invested capital and growth. Cheap companies within an industry peer group often deserve to be cheap. When considering comparable-multiple valuation, investors must look beyond the business attributes (they sell stuff) and focus on the business circumstances (what returns they earn and growth they promise). Investors must carefully describe the economic profile of a company before they can intelligently start the valuation process. The same thinking applies to considering where stocks trade relative to their historical multiples. Life cycle theory tells us that the economic grim reaperthe commodity P/Ewill eventually greet all companies. Reversion to the mean implies that the commodity P/E exerts a gravitational pull on companies. That said, some companies figure out ways to structurally improve their returns on capitaloften through better balance sheet management.13 Historical P/Es do not help investors when a companys underlying economics change. Pundits often compare todays market P/E to past P/Es to divine whether or not the market is attractive. But for market averages to be comparable over time, the statistical properties of the population must be the same, or stationary. If the populations properties change over time, the data are nonstationary and all bets based on comparisons are off (or should be). Theoretical and empirical analysis of P/E ratios for the market suggests nonstationarity. Research of the past 125 years shows no statistically significant relationship between a P/E ratio at the beginning of the year and the subsequent 12- and 24-month returns. 14 The three main drivers of P/E nonstationarity for the market are the role of taxes and inflation; changes in the composition of the market; and shifts in the equity risk premium. 15

Page 6

Legg Mason Capital Management

The Expectations Approach One important point about this discussion: a companys stock price reflects a set of expectations about future financial performance. Low multiples generally reflect low (and justified) expectations and high multiples often indicate lofty expectations. An investor must anticipate expectation revisions. There are three steps to an expectations process. The first is to understand where expectations are today. Using a metaphor of a high jumpers likely success, this step determines the bars current height. If a company trades at a P/E above the commodity level, this assessment requires understanding the markets expectations for future return on invested capital, growth, and sustainability. Reverse engineering the stock price using a discounted cash flow model and consensus estimates is a good start. The second step is to determine how high the company can jump, or whether or not it is likely to exceed embedded expectations. This requires financial and strategic analysis. The final step is to make a buy, sell, or hold decision. Using a range of possible value outcomes and probabilities, an investor can calculate an expected value. Stocks trading at a discount to expected value are attractive, but the amount of the price discount to expected value and how long it takes for the value-to-price disparity to close ultimately determine excess returns. Value Investing and Value Creation When taken together, the value equation and the expectations approach make it clear why distinctions between value and growth investing are vacuous. Growth considered in isolation is indeterminate; an investor needs first to understand returns on invested capital. Expectations investing shows that what matters is not the absolute level of todays implied performance but rather where expectations will move. Classic value investors seek to arbitrage the difference between price and value. Often, they find their portfolios filled with justifiably cheap business that earn returns on capital near or below the cost of capital. We call these investors value traders. Ideally, value investors can find businesses with prices below value where the value will increase over time. This value increase comes as management successfully deploys capital at attractive returns and fends off the migration to the commodity P/E multiple. In this case, value creation for the investor compounds in two ways as the price to value gap narrows and as value grows.

Page 7

Legg Mason Capital Management

Summary Often what passes as valuation is the simple application of historical or relative multiples without full consideration of what a multiple means. Breaking down a valuation, as Miller and Modigliani did over 40 years ago, provides critical insights into what drives corporate value and hence valuations. Here are three of the errors we see most often: Focus on growth instead of returns. Both companies and investors generally consider growth an absolute good. Indeed, a substantial part of corporate executive compensation still hinges on earnings growth. The value equation shows clearly that growth is only an amplifierand can work for you or against you. Comparisons of companies based on their attributes (industry) instead of their circumstances (economic model). History is filled with companies that lost out to disruptive technologies. In many of those cases, the disruptive business learned to make money at a lower price pointoften as the result of more efficient balance sheet utilization. Comparing companies to industry peers frequently doesnt properly take into consideration returns and growth. Use of historical multiples. Historical multiples only apply when the economic circumstances (both macro and micro) are unchanged. As a company loses its economic franchise, its multiple will drift to the commodity level. In some instances, companies do figure out ways to improve returns or extend their period of excess returns, both leading to higher multiples than history would indicate.

Page 8

Legg Mason Capital Management

Appendix A
This short appendix makes two main points. First, valuation does not change in the face of rising rates unless the real (inflation-adjusted) discount rate and/or the real growth rate change. Substantial evidence suggests that these rates are very sticky over time in the aggregate. But investors must exercise judgment for an individual company. Second, investors tend to be very poor at adjusting their earnings growth rates to reflect inflation. Generally, investors take recent inflation rates and extrapolate them. Modigliani and Cohn forcefully argued this point 25 years ago, and Campbell and Vuolteenaho reiterate the point in a current paper. 16 Valuation and rate changes A logical starting point is to show that you get the same value if you model a financial asset on a real or nominal basis. Assume a firm and the following conditions: 17 Distributable earnings = Growth next 3 years (real) = Growth beyond 3 years (real) = Ten-year note = Equity risk premium = Beta = Expected inflation = What are the growth rates? Real First 3 years 5% 3 years + 3% What are the discount rates? Real (1.0875/1.03) 1 = 5.58% What are the cash flows? Year 1 2 3 Real Cash flow 105.0 110.3 115.8 $100 5% 3% 4.75% 4% 1.0 3% Nominal (1.05)*(1.03) 1 = 8.2% (1.03)*(1.03) 1 = 6.1% Nominal 4.75% + 1*(4.0%) = 8.75% Nominal Cash flow 108.2 117.0 126.5

What are the terminal values? Real 115.8*(1.03)/(0.558 0.03) = $4,617 What are the present values? Real 105/1.0558 + 110.3/(1.0558)2 + (115.8 + 4,617)/(1.0558)3 = $4,219

Nominal 126.5*(1.061)/(.0875 - .0601) = $5,045

Nominal 108.2/1.0875 + 117.0/(1+1.0875) 2 + (126.5 + 5,045)/(1+1.0875)3 = $4,219

Page 9

Legg Mason Capital Management

The main message is the importance of consistency in matching cash flows with discount rates when you model. Provided you feel real growth rates havent changed (i.e., the company can price its goods or services at the rate of inflation) a change in nominal discount rates due to inflation will not change value. Broadly speaking, rising rates are consistent with rising earnings growth rates. The Inflation Illusion In 1979, Modigliani and Cohn claimed stock market investors are subject to inflation illusion. Specifically, they argued stock market investors fail to understand the effect of inflation on nominal growth rates. Rather, they extrapolate historical nominal growth rates even in times of changing inflation. Recognizing this tendency, investors must remain very careful to make sure cash flows and discount rates are consistent.

[email protected]

Page 10

Legg Mason Capital Management

Appendix B
Level of investment and return on invested capital form two of the three drivers of future value. Whats going on today is interesting on both counts. First, lets look at the investment side. Corporate investment spending today is well below historical levels as a percentage of cash flow and depreciation. (See Exhibit 4.) Three factors help explain why corporate chieftains currently have tight purse strings. Exhibit 4: Purse Strings Remain Tight Exhibit 2: Purse Strings Remain Tight Public Companies
% 100
Capital Spending-toGross Cash Flow

Rat ios of Capital Spending-to-Gross Cash Flow and Depreciation 1 1955 Through M id-December 2004

80 70 60 50 40 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 -to-Gross Cash Flow

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0

-to-Depreciation

Source: Corporate Reports, Empirical Research Partners Analysis. 1 Largest 1,500 stocks, data smoothed on a trailing six-month basis. Copyrighted material, used by permission.

First, there was a period of overinvestment in the late 1990s. Second, the government has placed regulatory burden on companies, prompting companies to turn inward. Finally, there has been a lack of confidence due to economic, regulatory, and political uncertainty. As a result, corporate balance sheets are the strongest they have ever been. There is evidence, though, that executives are more actively deploying capital, including an uptick in M&A activity. Next, consider the return side. Following a return boom in the late 1990s and a bust in the early 2000s, corporate returns are now approaching all-time highs again. (See Exhibit 5.) These high returns likely reflect the judicious or perhaps timidcapital deployment in recent years. Corporate America is earning well above its cost of capital.

Page 11

Legg Mason Capital Management

Capital Spending-toDepreciation

90

2.8 2.6

Exhibit 5: Aggregate CFROIs and Discount Rates (1986-2004)

Aggregate CFROIs & Discount Rates
usa 9 8 7 6 5 4 3 2 1 0
198606 198612 198706 198712 198806 198902 198908 199002 199008 199102 199108 199202 199208 199302 199308 199402 199408 199502 199508 199602 199608 199702 199708 199802 199808 199902 199908 200002 200008 200102 200108 200202 200208 200302 200308 200402 200408

Gross Investm ent Weighted CFROI Used in Valuation

Enterprise Value Weighted Discount Rate

Source: CSFB HOLT. Copyrighted material, used by permission.

Whats next? Companies are now turning their attention to capital deployment. Possible outcomes include a boost in M&A activity, a further recovery in investment spending, and a more concerted effort to return cash to shareholders via dividends and share repurchases.

Page 12

Legg Mason Capital Management

Endnotes
http://www.jfklibrary.org/j061162.htm. Paul Asquith, Michael B. Mikhail, and Andrea S. Au, Information Content of Equity Analyst Reports, National Bureau of Economic Research Working Paper 9246, October 2002. 3 Merton H. Miller and Franco Modigliani, "Dividend Policy, Growth, and the Valuation of Shares, The Journal of Business, 34, October 1961. The authors first express this concept in equation 12. Equation 22b in footnote 15 provides the simplest version of this equation. 4 This version of the steady state assumes that future cash flows are constant in nominal terms but decrease in real terms. For a full discussion of the difference between the perpetuity and perpetuity-with-inflation methods, see Alfred Rappaport and Michael J. Mauboussin, Expectations Investing (Boston, MA: Harvard Business School Press, 2001), 36-38. Please see the appendix for a brief discussion of how rising rates affect valuation. 5 See Bruce C. N. Greenwald et al., Value Investing (New York: John Wiley & Sons, 2001), 71-87. 6 G. Bennett Stewart, The Quest for Value (New York: HarperBusiness, 1991), 294-297. Also, see Dennis C. Mueller, Profits in the Long Run (Cambridge, UK: Cambridge University Press, 1986), 17-27. 7 Myron J. Gordon, The Investment, Financing, and Valuation of the Corporation (Homewood, IL: Irwin, 1962). 8 Rappaport and Mauboussin, 15-16. 9 Return on invested capital here = NOPAT T2-T1/Invested Capital T1-T0. 10 For this exhibit, we screened for earnings growth from fiscal year 1 to fiscal year 2 and CFROI for fiscal year 1. We sorted the sample by CFROI quartiles, and then sorted by EPS growth. The P/Es in the table are the median for sample of companies that fall into each cell. 11 See Palepu, Healy and Bernard, Business Analysis and Valuation (Cincinnati, OH: Southwestern College Publishing, 2000), 10-6; Pankaj Ghemawat, Commitment: The Dynamic of Strategy (New York: The Free Press, 1991), 82; Bartley J. Madden, CFROI Valuation (Oxford, UK: Butterworth-Heinemann, 1999): and Mauboussin, Schay, and McCarthy, Competitive Advantage Period: At the Intersection of Finance and Competitive Strategy, Frontiers of Finance, Credit Suisse First Boston Equity Research, October 4, 2001. 12 Mark Veverka, Super-Seller Amazon Rates a Sell, Barrons Online, November 29, 2004. 13 For an example of improved balance sheet efficiency, see the McDonalds example in Rappaport and Mauboussin, 148-149. For a fuller discussion of increasing returns, see Michael J. Mauboussin, Exploring Network Economics, Mauboussin on Strategy, October 11, 2004. 14 Kenneth L. Fisher and Meir Statman, Cognitive Biases in Market Forecasts: The Frailty of Forecasting, Journal of Portfolio Management, Fall 2000, 72-81. 15 Michael J. Mauboussin and Kristen Bartholdson, Leaving the (non)Station(arity), The Consilient Observer, May 6, 2003. 16 Franco Modigliani and Richard Cohn, Inflation, rational valuation, and the market, Financial Analysts Journal, 1979; John Y. Campbell and Tuomo Vuolteenaho, Inflation and Illusion and Stock Prices, NBER Working Paper, December 15, 2003. See http://kuznets.fas.harvard.edu/~campbell/papers/inflation20031215b.pdf. 17 We adapted this from Aswath Damodaran, Damodaran on Valuation (New York: John Wiley & Sons, 1994), 56-58.
2 1

Page 13

Legg Mason Capital Management

References Articles
Asquith, Paul, Michael B. Mikhail, and Andrea S. Au, Information Content of Equity Analyst Reports, National Bureau of Economic Research Working Paper 9246, October 2002. Campbell, John Y., and Tuomo Vuolteenaho, Inflation and Illusion and Stock Prices, NBER Working Paper, December 15, 2003. Fisher, Kenneth L., and Meir Statman, Cognitive Biases in Market Forecasts: The Frailty of Forecasting, Journal of Portfolio Management, Fall 2000. Mauboussin, Michael J., Alexander Schay, and Patrick McCarthy, Competitive Advantage Period: At the Intersection of Finance and Competitive Strategy, Frontiers of Finance, Credit Suisse First Boston Equity Research, October 4, 2001. Mauboussin, Michael J., and Kristen Bartholdson, Leaving the (non)Station(arity), The Consilient Observer, May 6, 2003. Mauboussin, Michael J., Exploring Network Economics, Mauboussin on Strategy, October 11, 2004. Miller, Merton H., and Franco Modigliani, "Dividend Policy, Growth, and the Valuation of Shares, The Journal of Business, 34, October 1961. Modigliani, Franco, and Richard Cohn, Inflation, rational valuation, and the market, Financial Analysts Journal, 1979. Veverka, Mark, Super-Seller Amazon Rates a Sell, Barrons Online, November 29, 2004.

Page 14

Legg Mason Capital Management

Books
Damodaran, Aswath, Damodaran on Valuation (New York: John Wiley & Sons, 1994). Ghemawat, Pankaj ,Commitment: The Dynamic of Strategy (New York: The Free Press, 1991). Gordon, Myron J., The Investment, Financing, and Valuation of the Corporation (Homewood, IL: Irwin, 1962). Greenwald, Bruce C. N., et al., Value Investing (New York: John Wiley & Sons, 2001). Leibowitz, Martin L., Franchise Value: A Modern Approach to Security Analysis (New York: John Wiley & Sons, 2004). Madden, Bartley J., CFROI Valuation (Oxford, UK: Butterworth-Heinemann, 1999). Mueller, Dennis C., Profits in the Long Run (Cambridge, UK: Cambridge University Press, 1986). Palepu, Krishna G., Paul M. Healy, and Victor L. Bernard, Business Analysis and Valuation (Cincinnati, OH: Southwestern College Publishing, 2000). Rappaport, Alfred, and Michael J. Mauboussin, Expectations Investing (Boston, MA: Harvard Business School Press, 2001). Stewart, G. Bennett, The Quest for Value (New York: HarperBusiness, 1991).

Page 15

Legg Mason Capital Management

Legg Mason Capital Management (LMCM:) is comprised of (i) Legg Mason Capital Management, Inc., (ii) Legg Mason Funds Management, Inc., and (iii) LMM LLC. The views expressed in this commentary reflect those of LMCM as of the date of this commentary. These views are subject to change at any time based on market or other conditions, and LMCM disclaims any responsibility to update such views. These views may not be relied upon as investment advice and, because investment decisions for clients of LMCM are based on numerous factors, may not be relied upon as an indication of trading intent on behalf of the firm. The information provided in this commentary should not be considered a recommendation by LMCM or any of its affiliates to purchase or sell any security. To the extent specific securities are mentioned in the commentary, they have been selected by the author on an objective basis to illustrate views expressed in the commentary. If specific securities are mentioned, they do not represent all of the securities purchased, sold or recommended for clients of LMCM and it should not be assumed that investments in such securities have been or will be profitable. There is no assurance that any security mentioned in the commentary has ever been, or will in the future be, recommended to clients of LMCM. Employees of LMCM and its affiliates may own securities referenced herein.

Page 16

Legg Mason Capital Management