THE BOEING 7E7

Teaching Note

Synopsis and Objectives

In 2003, the Boeing Company announced plans to build a new “super-efficient” commercial jet called the “7E7” or “Dreamliner.” This was a “bet the farm” gamble by Boeing, similar in magnitude to its earlier introductions of the 747 and 777 airliners. The technological superiority of the new airframe, as well as the fact that it would penetrate a rapidly growing market segment, were arguments for approval of the project. On the other hand, the current market for commercial airplanes was depressed because of terrorism risks, war, and SARS, a contagious illness that resulted in global travel warnings. Boeing’s board of directors would need to weigh those considerations before granting final approval to proceed with the project.

The task for students is to evaluate the 7E7 project against a financial standard, the investors’ required returns. The case gives internal rates of return (IRR) for the 7E7 project under base-case and alternative forecasts. The students must estimate a weighted-average cost of capital (WACC) for Boeing’s commercial-aircraft business segment in order to evaluate the IRRs. As a result of that analysis, the students identify the key value drivers and distinguish, on a qualitative basis, the key gambles that Boeing is making.

The general objective of this case is to exercise students’ skills in estimating a weighted-average cost of capital and cost of equity. The need for students to estimate a segment WACC draws out their abilities to critique different estimates of beta and to manipulate the levered-beta formulas. Boeing competes in both the commercial aircraft and the defense business. Thus, deriving the appropriate benchmark WACC for the 7E7 project requires isolating the commercial aircraft component from Boeing’s overall corporate WACC. In doing so, students engage the concept of value additivity.

Suggested Questions for Advance Assignment to Students

1. What is an appropriate required rate of return against which to evaluate the prospective IRRs from the Boeing 7E7?

a. Please use the capital asset pricing model to estimate the cost of equity. At the date of the case, the 74-year equity market risk premium (EMRP) was estimated to be ___. Which beta and risk-free rate did you use? Why?

b. When you used the capital asset pricing model, which risk-premium and risk-free rate did you use? Why?

c. Which capital-structure weights did you use? Why?

2. Judged against your WACC, how attractive is the Boeing 7E7 project?

a. Under what circumstances is the project economically attractive?

b. What does sensitivity analysis (your own and/or that shown in the case) reveal about the nature of Boeing’s gamble on the 7E7?

3. Should the board approve the 7E7?

Supporting Excel Spreadsheets

Student analysis of the case is supported by the spreadsheet file, “Case_16.xls,” available from Darden Business Publishing (sales@dardenbusinesspublishing.com). Instructor analysis is supported by “TN_16.xls.” It is a condition of accessing this file that you agree not to share the contents of the instructor file with students.

Hypothetical Teaching Plan

The following questions offer an outline for discussion leadership. These can be easily condensed or expanded to meet a discussion time as short as 80 minutes and as long as 4 hours, depending on the depth to which the instructor wishes to address the issues. 1. Why is Boeing contemplating the launch of the 7E7 project? Is this a good time to do so?

This opening summarizes the basic facts, broad motives for the project, and the risks. The objective for this part of the discussion is to set the tone for the case discussion, especially the need to prepare a recommendation for Boeing’s board of directors.

2. Should Boeing’s Board approve the 7E7?

The instructor can take the students’ vote and then solicit a few opinions and summarize key arguments on the chalkboard. One can ask specific students for their estimate of the cost of capital and write it next to their yes or no vote. The students will show some division of opinion, but in our classroom experience, they tended to lean toward approval. The instructor could easily take a “devil’s advocate” approach toward the prevailing sentiment and invite defense of the students’ opinions. In any event, acknowledgment of a range of estimates builds drama and confirms that there are competing views among students—this motivates a detailed discussion of the financial analysis.

3. How would we know if the 7E7 project will create value?

Students may be familiar with the classic NPV criterion. This case invites them to focus on the internal rate of return (IRR). If the IRR is greater than the project cost of capital, the 7E7 is a positive net present value project.[1] A discussion of why this is true provides a solid “big picture” foundation for the case decision. The project IRRs are presented in case Exhibit 9. Therefore, the focus of student analysis should be on determining the benchmark against which to evaluate the IRRs. Thus, this part of the discussion helps to motivate the analysis of WACC. Some students may have voted in a manner that contradicts the IRR versus the cost of capital decision rule. This sets up the next question.

4. Okay, let’s examine the details of how to estimate the WACC. Let’s go step-by-step. Where shall we get started?

The instructor can ask one student (or team of students) to walk the class through the detailed calculations. The presenter(s) may make one or more errors or adopt some controversial practice in making the calculations. This presents a strategic teaching question for the instructor: interrupt and correct at the first mistake, or do it after they are done? The latter is preferable if you are in a discussion-based learning environment. Thus, the instructor could keep a careful record of the students’ calculations on the chalkboard,[2] let them finish, and then turn to the class and ask, “Do you all agree? Why not?” Challenges and debates over specific issues will follow. At this point, it is not important to judge the quality of the assumptions (they can be good or bad). The task of the instructor is to work through the WACC calculation in an orderly manner. A number of thought-provoking questions about techniques can lead to rich classroom discussions. By the close of this segment of the discussion, the instructor should aim to have on the chalkboard a finished WACC calculation that highlights aspects about which there may be some disagreement.

5. The weighted-average cost of capital is a simple formula. Yet it seems that reasonable people can disagree about the estimates. How can there be such a range of cost of capital estimates? Please summarize the issues.

This segment asks the students to summarize the thorny questions of estimating WACC and to contrast the issues in practice with the underlying simplicity of the weighted-average cost of capital formula and the capital asset pricing model. The reason for varying estimates is that students will have made different assumptions as they apply the formula. This will be apparent on the chalkboard if the instructor has faithfully recorded the assumptions of the main presenter(s) and any alternate assumptions that may have surfaced. The next step is to do sensitivity analysis on each assumption. If the sensitivity analysis reveals that the decision outcome is significantly impacted by the assumption, only then is it worth spending more time, energy, and dollars on improving the quality of the assumption. This is another judgment part of financial analysis and decision-making. A good analyst will make the board aware of critical assumptions, so that it can, in turn, make more informed and better decisions.

6. Have we thought of everything? Is there anything else the board of directors should consider in assessing the financial appeal of this project? Why might the board vote “yes” on the 7E7, when the cost of capital estimate is greater than the IRR? Why might the Board vote “no” if the cost of capital estimate is less than the IRR?

This segment of the discussion should aim to introduce the basic idea that determinate cash flow forecasts do not capture contingent values. Real option valuation is beyond the scope of this case, but students should be reminded that a large capital project such as the 7E7 is probably riddled with rights that can enhance the value of the project—rights to new yet-to-be discovered technology, rights to grow and/or enter new markets, and rights to cross-pollinate other projects’ new intellectual property from the 7E7. In addition, the 7E7 might create intangible value for Boeing’s brand and strategic value for cross-selling. The potential significance of qualitative issues can be an eye-opening point for some students who consider finance to be a “numbers” class. In the end, the board of directors must weigh both the determinate cash flow values and the contingent and intangible values in the project.

7. What should the board do?

The instructor could bring closure to the discussion with another student vote. Making careful note of students who may have changed their minds, the instructor could invite one or two to comment on what they learned. Airing the reasoning behind some of the vote changes will reinforce the learning objectives underlying the decision-making process. The instructor could close the class with a brief commentary on development of the 7E7 project since the date of the case. The epilogue presented in this note recounts the board’s decision to proceed with the 7E7 project. Further developments may be gleaned from Boeing’s Web site and others.

Case Analysis: Estimation of the Project-Specific WACC

Launch and timing of the 7E7

The motives for the project are laudable: the 7E7 is entering a good growth segment of the industry. Higher performance and fuel efficiency will position Boeing favorably in the market, and perhaps, take back some market share from competitors. R&D on this project may create inventions that will prove to be valuable to other Boeing products. At the same time, the consequences of error are staggering: this is a bet-the-ranch kind of investment.

But, as most entertainers know, timing is everything. Here, the timing could not be worse: war, airline-focused terrorism, SARS, and the weak financial condition of airlines all challenge the approval of the project. “Why now?” is a question that the board must answer. In part, the answer depends on the long development cycle (four years) and very long product lifecycle (20 years). The board is making a bet less on conditions that prevail today than on conditions that are expected to prevail many years into the future.

Calculating WACC

The instructor could use the formulas for WACC and the capital asset pricing model as a format for organizing the calculation section of the discussion.

1. WACC = (% debt)(rd)(1 – tc) + (% equity)(re)

where: rd = required rate of return of debt

re = required rate of return on equity (cost of equity)

tc = marginal effective corporate tax rate

% debt = debt divided by sum of debt plus equity, measured at market value

% equity = equity divided by sum of debt plus equity, measured at market value. Market value of equity is simply share price times number of shares. Many analysts simply approximate market value of debt with its face or book value. In most instances, this is not an unreasonable practice: for floating-rate debt issues (such as bank loans) market value will equal book value; for fixed-rate issues, book value may be a close substitute for market value, where the bond was issued at face value, and the credit quality of the issuer and the general level of interest rates have not changed since date of issue.

2. Calculating the cost of equity (re)

Per the CAPM, re = rf + βe (rm – rf)

Where rf is the risk-free rate, rm – rf is the market-risk premium and βe is the beta of equity.

One can leave the formulas on the board so that students are able to see what the class is working toward, as a simple organizing framework; but this simplicity is deceptive. The theory says relatively little about the practical application of those models. Reasoning from first principles, one can make good headway in estimating the cost of capital. The kinds of issues that will occupy the student include the following:

Choice of models: Students may wonder why the capital asset pricing model is not used to estimate the firm’s cost of capital directly. The answer is that it can be used in such a fashion, but in practice is almost never used for that purpose. A true asset beta is unobservable because the items on the left-hand side of the balance sheet are ordinarily not traded in liquid markets. In theory, one could derive a beta for debt and a beta for equity and then weight them according to the market values of their securities—this should give an estimate for the firm’s asset beta. Alternatively, one could unlever the equity beta to get an asset beta. Inserting the asset beta into the CAPM would, in theory, yield an estimate of the firm’s cost of capital. But these latter two methods invite estimation error. A survey of “best practice” firms,[3] however, revealed a total preference for using the WACC formula and for restricting CAPM to estimating the cost of equity.

Beta for the 7E7 project: The tendency of novices will be to use Boeing’s equity beta from case Exhibit 10 as an input to the CAPM. This would be inappropriate since this would assume that Boeing’s commercial aircraft risk is equal to Boeing’s firm risk. However, Boeing’s firm risk is a blend of both commercial and defense risk. Somehow, for the cost of capital to be meaningful we have to back out the commercial risk from the defense risk and use a beta reflecting commercial risk in the CAPM. The class could discuss whether one would expect a commercial beta to be higher or lower than a defense beta and why—a strong case could be made that defense would have a lower beta, reflecting the zero default risk of Boeing’s government clients and the long cost-plus contracts. It is useful to exercise students’ intuition about risk.

Estimates of beta: Case Exhibit 10 poses two basic choices: first, what to use as a proxy for the market when regressing equity returns on market returns; and second, over what period of time the beta should be measured.

Ideally, beta should be derived by regressing Boeing’s equity return against the return on the basket of global assets (i.e., the market return). Measuring the return on the global asset portfolio is, as a practical matter, proxied by returns on equity market indexes. In bridging theory with practice, the greater the value captured by the index, the more closely it matches the theory. In case Exhibit 10, students have a choice between the New York Stock Exchange (NYSE) composite index and the S&P 500 index. Of those two, the NYSE is a broader and higher value index. This provides a beta range between 1.00 and 1.62.

Next, students have to choose whether to estimate beta across the 60-day, 21-month, or 5-year period. Beta is the market’s instantaneous perception of risk and moves on a second by second basis just like stock prices. However, since this is unobservable, we are forced to use historic data to estimate the market’s perception of risk today. The question therefore is what historic period best reflects the risk of the commercial aircraft industry as perceived by the market at the time of the decision?

Using a 60-day beta regression includes the trading dates between March 20 and June 16, 2003. This time period includes the Iraq war as well as the peak of the SARS travel warnings. The 21-month beta runs from September 17, 2001, and therefore includes significantly more terrorism risk. The 60-month beta dilutes terrorism, war, and SARS risk by going as far back as June 16, 1998. Students will want to have a discussion of what time period to use as it relates to the commercial-aircraft industry, however, this may be a good time to emphasize the point that such a discussion is only productive after doing sensitivity analysis with the different betas. For now, this is a benchmark case, and it is only important to keep track of all assumptions. For purposes of illustration, this note uses the 60-day beta of 1.62.

Calculating a beta for Boeing’s commercial division: Exhibit TN1 shows pictorially the underlying logic in calculating the commercial beta. Step 1 is to use the unlevered beta formula to determine what the beta would have been if Boeing had no debt.[4] If Boeing had no debt, then the unlevered beta on the right-hand side of the balance sheet would be equal to the beta on the left-hand side: Boeing’s asset beta. In step 2, the hypothesis is that Boeing’s asset beta is a weighted average of its commercial and defense beta.

Exhibit TN2 shows a brief derivation of the unlevered beta formula. It may be appropriate to discuss the assumptions underlying the formula, and whether or not those assumptions are reasonable as they pertain to Boeing. The formula assumes riskless debt and that the only impact of debt on the value of the firm is the corporate debt tax shield. The formula for the debt tax shield (tcD) assumes permanent debt and that the company will be permanently profitable enough to use the tax shields. Exhibit TN3 shows Boeing’s unlevered beta with inputs from case Exhibit 10, using the 60-day beta as measured against the NYSE as the levered equity beta. Exhibit TN4 illustrates the idea that the unlevered beta of 1.21 is a weighted average of Boeing’s commercial and defense division betas. To solve for the commercial Boeing beta, it is necessary to determine a defense beta and to decide how to estimate the percentage division weights. This process can entail some interesting questions that result in rich classroom discussion, around issues such as:

Weights for defense and commercial segments: To bridge theory with practice, in answering this, the template is the same as before. First, what should the weights be in theory? In theory, they should be the market value of Boeing’s defense and commercial assets. Case Exhibit 1 shows Boeing’s revenues, operating profits, and identifiable assets broken up by the commercial and defense segments. Each of those measures has advantages and disadvantages as a proxy for the market value of assets. A discussion of the merits of each measure will again emphasize the importance of applying judgment when bridging theory with practice. As a benchmark case, we use the percentage of revenues as weights (also shown in case Exhibit 10).

Betas for defense and their unlevering: Case Exhibit 10 shows defense betas for Northrop Grumman and Lockheed Martin, which both have over 90% of their revenues coming from defense. If we assume that they are 100% in the defense business then we can use their betas to proxy for defense. We should use the 60-day NYSE betas to be consistent with the earlier steps in the example. Throughout this case discussion, students may mix assumptions. Internal consistency is important. Hence, the relevant betas are 0.34 and 0.27. Some students may be tempted to average these levered betas—but doing so would reflect the financing decisions of the peer firms and not that of Boeing. What the calculation requires is the unlevered beta of the defense segment. Therefore, the betas of Lockheed and Northrop must be unlevered and then averaged as shown in Exhibit TN5. Two points emerge. First, it makes sense to average Lockheed Martin and Northrop Grumman betas of 0.29 and 0.21 respectively—this simply draws on the greater information embedded in two (rather than, say, one) observations. The average is 0.25. The two peer betas are less than 0.1 apart, relatively tightly clustered. Second, based on an earlier discussion, students should already have expected the defense business to have a lower beta than the commercial business.

Beta for commercial segment: As shown in Exhibit TN6, the commercial beta is 2.03, greater than the defense segment betas as expected. At this point, the 2.03 commercial beta is unlevered. If you applied this beta to the CAPM, you would be obtaining a discount rate for the commercial aircraft business and there would be no need to use the WACC formula. While this is technically correct, you would not be using market provided information on debt at the time of the case. Hence, a superior approach would be to relever the 2.03 to obtain a beta to use in CAPM to calculate the cost of equity based on commercial risk. Exhibit TN7 contains commercial betas consistent with a range of assumptions the instructor might encounter in class—the range spans the values 1.04 to 3.31.

Calculating cost of equity: Exhibit TN8 shows the remaining inputs and calculations needed to estimate Boeing’s commercial cost of equity. Capital-asset pricing theory gives no guidance about which risk-free rate of return to use. The case includes both a three-month and a 30-year choice of Treasury rates in June 2003. Some students will point out that a longer-term rate is warranted to match the length of the project. Others may point out that default is less likely in the shorter term, so a short-term rate more closely matches the theory. My response is that both arguments have merit and remain unresolved theoretical territory. However, sensitivity analysis typically reveals that this is not an important pragmatic issue. The illustrated calculation of WACC in this note uses as the risk-free rate the three-month T-bill rate in June 2003 of 0.85%, which results in a WACC of 16.7%. The WACC would change to 15.6% if the 30-year T-bond rate were used (based on the arithmetic equity risk premium).

In theory, the CAPM reflects the market’s assessment of the premium today. Similar to beta, current practice uses historic data to make this estimate. Hence, the time period over which the average is measured can be subject to meaningful judgment. For example, does the last 20 years or last 74 years better reflect the market’s assessment of the premium today?

Exhibit TN9 gives the range of equity cost estimates consistent with various assumptions. The largest variation in estimates is due to assumptions about beta and the equity market risk premium. Estimates span the values from 10.19% to 28.65%.

Calculating the cost of debt (rd): Case Exhibit 11 provides the information needed to calculate Boeing’s cost of debt. The approach is to use a market value weighted average of yields to maturity of Boeing’s different debt issues as of June 2003. Exhibit TN10 shows the calculations. A caveat in the calculation is that case Exhibit 11 does not include all of Boeing’s debt. It is missing short-term debt as well as debt issued by Boeing’s financing subsidiary: Boeing Capital Corporation. In theory, a weighted average of all debt reflects the debt component risk on the left-hand side of the balance sheet. In practice, the yields of debt issued by Boeing Capital would be accessible information and this compromise in the calculation would not be necessary. Given this caveat, some interesting questions can be discussed:

1. Is debt not also subject to commercial and defense risk, and should we therefore not try to back out the commercial risk component?

While this point is perfectly valid, it is much less of an issue than equity since debt is a senior security.

2. Why not simply use the CAPM to estimate the cost of debt?

While the beta of each debt issue is possible to estimate if it is traded publicly, and while CAPM is not restricted to equity instruments in theory, solving for a bond’s yield to maturity (YTM) using existing market data avoids all the assumptions and weaknesses inherent in CAPM. The only reason we are forced to use a model like CAPM in estimating equity is that, unlike a bond instrument where coupon and principal payments are known with a fair degree of certainty, we do not know what cash flows to expect over the potential infinite life of the equity security. 3. Should you calculate a weighted average of all debt or a weighted average of long-term debt with maturities that match the length of the project?

Theory tells us that a weighted average of all debt risk is a function of the asset risk on the left-hand side of the balance sheet. However, it is a matter of debate whether one should further sort debt risk by matching debt maturity to the length of the project. While picking a debt instrument whose maturity matches the length of the project may be pleasingly intuitive, a counterargument would be to ask: what if Boeing only had short-term debt? In the case of Boeing, sensitivity analysis reveals little difference between the two approaches. A weighted average of all debt maturing on or after the year 2031 (maturities ranging from 27–39 years) results in a cost of debt of 6.03%. This compares to 5.33% using a weighted average of all available debt information. This would change the WACC from the base case of 16.7% to 16.9%.

Tax rate: The theory requires an analyst to use the marginal, expected tax rate. Some students may use historical, average tax rates (which they derive by dividing annual tax expense by pretax profits). As the case notes, Boeing’s tax exposure has been lower in the past than it is expected to be in the future; using the marginal effective tax rate of 35% is, therefore, more appropriate.

Capital-structure weights: The theory requires an analyst to use market value, not book value, weights. Case Exhibit 10 provides the market value debt/equity ratio for Boeing as 0.525. From this, you can derive the percentage debt and equity to be 34.4% and 65.6% respectively.

The WACC estimate: With all the assumptions given in this illustration, Exhibit TN11 shows the base-case estimate of the WACC to be 16.7%, greater than the base-case IRR scenario of 15.7%. This appears to be an economically unattractive project. The only way to establish a level of confidence with this conclusion is to test WACC’s sensitivity to variations in the underlying assumptions. Exhibit TN12 presents a range of WACC estimates consistent with various assumptions that the instructor might encounter. The revelation is that the vast majority of the estimates are smaller than 15.7%. A key difference has to do with the term of the beta used in the estimate.

The 60-day beta, for example, gives heavy weight to the Iraq war and SARS. If you believe that SARS and war will have an ongoing and significant risk to the commercial-aircraft industry, then this should be your beta of choice. Conversely, you could argue that the 21-month beta (beginning September 2001) places more weight on the broader history of two wars (Afghanistan and Iraq) and the general risk of terrorism. The 60-month beta dilutes impact of terrorism, war, and SARS risk.

Need for a “view”: Plainly, the practitioner needs to have a view about political and economic fundamentals in the future. How one gains such a view is beyond the scope of this case discussion, although the instructor could point out that numerous courses in a typical business degree program arm the analyst with tools and concepts useful in forming a view. History is one guide to what might happen in the future. Only 12 years earlier, Boeing committed to build the 777 aircraft in the face of a similar economic and political environment. Though history is never a perfect guide, it lends more insight than random guesses or agnosticism. In the final analysis, the numerical analysis is not enough; it must be supported by some kind of outlook.

The Problem for the Board of Directors

This phase of the discussion, in effect, shifts in order to reflect on the limitations of DCF analysis of projects. In short, methods that focus on the time value of determinate cash flows do not capture contingent cash flows, and by its very focus on the “project,” the forecast of cash flows may ignore other economic benefits outside of the 7E7 project, or in the realm of intangible value. The wise analyst understands that DCF likely tells only part of the story. The wise board of directors will ruminate over the range of effects that DCF might miss. These other factors might compel a board to approve a project whose IRR did not exceed its WACC, or alternatively, to reject a project when the IRR did exceed the WACC. Issues specific to this case that students might offer include these:

1. Strategic options: Perhaps this project creates rights or options; perhaps not undertaking this project would limit or shorten Boeing’s strategic flexibility elsewhere in the industry. Airbus has already overtaken Boeing in this industry by many measures, and, perhaps without the 7E7, Boeing would essentially relinquish permanent and dominant control of this industry. If they ever tried to re-emerge as a commercial aircraft industry leader, would they have that ability?

2. Growth and R&D options: Rights to grow in new markets are valuable options: arguably, the 7E7 positions Boeing favorably with opportunities to grow in the regional jet market. Technological spin-offs, from the 7E7 project to other projects at Boeing, create technology options for the company—innovations in the use of composite materials and fuel-saving designs can position Boeing to compete more effectively against Airbus.

3. Related activities in other divisions: The new product line will augment Boeing’s aftermarket parts and service business. Unclear from the forecast is the extent to which this activity is captured in the 7E7 projections—but even if it is, perhaps there are scale economies in parts and service that the 7E7 will help other divisions of Boeing exploit.

The board should also scrutinize opportunities for effective risk management through operations that could reduce Boeing’s exposure on this project. For instance, case Exhibit 9 shows that the IRR is very sensitive to both development and production costs. If the percentage of COGS/sales goes from 80% (base case) to 84%, then the IRR plummets from 15.7% to 10.3%. This sensitivity suggests that innovations in manufacturing processes, purchasing, and assembly might have a very large beneficial impact on the profitability of the project—if they pay. Another form of risk management by the board would be to establish milestones at which progress on the project would be reviewed and changes made, if necessary. Tailoring managerial incentive systems to keep within the financial parameters is another form of risk management. Students may summon more examples of techniques that can limit Boeing’s downside on the 7E7.

Ultimately, Boeing is in the business of building airplanes. No doubt, the board and senior management will feel a strong impulse to approve the 7E7 project, as a statement about Boeing’s commitment to the commercial airplane business. The mission of the firm is an extremely important consideration in the review of a project such as this. But the financial staff must lend some discipline in pursuit of the mission. The 7E7 project is certainly not the only new airframe opportunity facing Boeing. Project analysis needs to establish the attractiveness of this project at this time.

In the end, the board would probably be reluctant to concede an attractive market segment to competitors, but would impose risk management conditions that would protect the firm. This is how the board proceeded.

Epilogue

In an unprecedented and highly risky move, in September 2003, the board voted to take orders for the 7E7 on a limited basis. Specifically the board provided permission to offer the 7E7 to Japan’s two largest airlines: All Nippon Airways and Japan Air Lines. It was considered particularly risky, however, to launch the 7E7 without a major U.S. customer. For example, when Boeing developed the 777, they worked closely with their U.S. customers, who in turn had significant influence on the plane’s design. While the Asian market was forecasted to be the largest growth market, the needs of the Japanese were short range, while the needs of the rest of the world were longer range. Boeing, therefore, was now considering three versions of the 7E7 to accommodate the Asian market and the markets outside Asia. Multiple models however typically drive up development costs, and the board had already stated that they would not back the 7E7 unless development costs were lower than what it took to develop the 777.

In December 2003, the board unanimously voted to give its sales force the authority to offer the 7E7 worldwide. Missing from this announcement, however, was any airline that had agreed to buy the plane so early in the process. To get updated information on Boeing’s aircraft orders, go to Boeing’s commercial aircraft Web site: www.boeing.com/commercial. On this page, one finds the option to obtain “commercial airplane info,” which provides ordering and delivery information.
Exhibit TN1

The Boeing 7E7

Boeing’s Commercial Beta (1)

Exhibit TN2

The Boeing 7E7

Unlevered Beta Derivation

Exhibit TN3

The Boeing 7E7

Sample Illustration: Boeing’s Unlevered Beta

Exhibit TN4

The Boeing 7E7

Boeing’s Commercial Beta

Exhibit TN5

The Boeing 7E7

Example: Unlevered Defense Beta of 0.25

Exhibit TN6

The Boeing 7E7

Example: Boeing’s Commercial Beta

Exhibit TN7

The Boeing 7E7

Range of Unlevered Beta Estimates for Commercial Segment

Exhibit TN8

The Boeing 7E7

Example: Commercial Cost of Equity

Exhibit TN9

The Boeing 7E7

Range of Cost of Equity Estimates

Exhibit TN10

The Boeing 7E7

Boeing’s Cost of Debt

Exhibit TN11

The Boeing 7E7

Example: WACC Estimate for Commercial Segment

Exhibit TN12

The Boeing 7E7

Range of WACC Estimates

Appendix

The Boeing 7E7

Background Comments on the Equity Market Risk Premium

The theory of capital asset pricing gives no guidance on its practical application. Finance theory says the equity market risk premium should equal the excess return expected by investors on the market portfolio relative to riskless assets. How one measures expected future returns on the market portfolio and on riskless assets are problems left to practitioners. A survey of “best practice” financial offices, by Bruner, Eades, Harris, and Higgins[5], on the estimation of the cost of capital, found that practitioners disagree most sharply on the question of the measurement of the equity market risk premium. Scholars, too, offer a wide range of views, as summarized in a roundtable discussion by Ivo Welch.[6] Because expected future returns are unobservable, “best practice” is to extrapolate historical returns into the future on the presumption that past experience heavily conditions future expectations. Where practitioners differ is in two dimensions:

• use of arithmetic versus geometric average historical equity returns

• choice of realized returns on T-bills versus T-bonds to proxy for the return on riskless assets.

The arithmetic mean return is the simple average of past returns. Assuming that the distribution of returns is stable over time and that periodic returns are independent of one another, the arithmetic return is the best estimator of expected return.[7] The geometric mean return is the IRR between a single outlay and one or more future receipts. It measures the compound rate of return investors earned over past periods and accurately portrays historical investment experience. Unless returns are the same each time period, the geometric average will always be less than the arithmetic average and the gap widens as returns become more volatile.[8]

The standard source for measured equity market risk premiums is Stocks, Bonds, Bills, and Inflation, an annual yearbook published by Ibbotson Associates. As noted on page two of this teaching note, Ibbotson’s estimates of EMRP from 1926 to 2003, offer a relatively wide
Appendix (continued)

range of values, depending on use of the geometric as opposed to the arithmetic mean equity return and on use of realized returns on T-bills as opposed to T-bonds.

Even wider variations in market risk premiums can arise when one changes the historical period for averaging. Extending U.S. stock experience back to 1802, Siegel[9] shows that historical market premia have changed over time and were typically lower in the pre-1926 period. Visual inspection of the Ibbotson estimates shows considerable variation in historical premia over different time periods and methods of calculation, even with data since 1926.

The survey by Bruner et alia found that a majority of texts and trade books support use of the arithmetic mean return over T-bills as the best surrogate for the equity market risk premium. Half of the financial advisors queried use a premium consistent with the arithmetic mean and T-bill returns, and many specifically mentioned use of the arithmetic mean. Corporate respondents, on the other hand, evidenced more diversity of opinion and tend to favor a lower market premium of about 6%.

This variety of practice should not come as a surprise since theory calls for a forward-looking risk premium, one that reflects current market sentiment and may change with market conditions. What is clear is that there is substantial variation as practitioners try to operationalize the theoretical call for a market risk premium.

As a practical matter, the best practice in the estimation of WACC is to adopt a “house view.” This view is driven by values and beliefs about markets and analysis. For instance, the analyst could consider these points:

• We do not calculate the cost of capital, we estimate it. This means that any WACC should be viewed as having a range of uncertainty surrounding it. We cannot eliminate the uncertainty; but best practice should aim to narrow the range. Uncertainty about EMRP is one source of uncertainty in our estimates of WACC. One response is to embrace this uncertainty in the analysis of projects, mainly by conducting sensitivity analysis of IRRs and NPVs, as driven by the range of WACCs.

• The choice of T-bills or T-bonds as the basis for EMRP depends on one’s view about the best way to impound inflation expectations in project analysis. Treasury instruments capture inflation expectations prevailing for the life of the instrument. A common practical response is to use a Treasury instrument that is contemporaneous with the life of the asset being valued. The Boeing 7E7 is a long-lived project and, therefore, could be gauged against the T-bond-based EMRP.

• The choice of geometric versus arithmetic is simply irresolvable. Both methods have desirable attributes and weaknesses. Here, thoughtful practitioners choose a method of calculation that yields results that seem reasonable.
-----------------------
[1] Of course, this assumes initial negative cash flows followed by positive cash flows. [2] This class is ideally suited for the chalkboard. PowerPoint or other digital media make it difficult to summarize the complete analytic process in plain view of the students and simply puts the success of the whole class in the hands of the presenters. The instructor needs to remain in control of the discussion—continuing to hold the chalk is the best way to do this. [3] See the study ‘Best Practices’ in “Estimating the Cost of Capital: Survey and Synthesis,” Financial Practice and Education (Spring/Summer 1998) by Robert F. Bruner, Kenneth M. Eades, Robert S. Harris, and Robert C. Higgins. [4] For a detailed exposition of unlevered beta, see Ross, Westerfield, and Jaffe, Corporate Finance, 6th ed., (2001), 481–484. [5] Robert F. Bruner, Kenneth M. Eades, Robert S. Harris, and Robert C. Higgins, “‘Best Practices’ in Estimating the Cost of Capital: Survey and Synthesis,” Financial Practice and Education (Spring/Summer 1998). [6] http://papers.ssrn.com/sol3/papers.cfm?abstract_id=234713 [7] Several studies have documented significant negative autocorrelation in returns—this violates one of the essential tenets of the arithmetic calculation, since if returns are not serially independent, the simple arithmetic mean of a distribution will not be its expected value. The autocorrelation findings are reported by: Fama and French (1986), Lo and MacKinlay (1988), and Poterba and Summers (1988). [8] For large samples of returns, the geometric average can be approximated as the arithmetic average minus one-half the variance of realized returns. Ignoring sample size adjustments, the variance of returns in the current example is 0.09 yielding an estimate of 0.10 " ((0.09) = 0.055 = 5.5% versus the actual 5.8% figure. Kritzman (1994) provides a− ((0.09) = 0.055 = 5.5% versus the actual 5.8% figure. Kritzman (1994) provides an interesting comparison of the two types of averages. [9] Jeremy J. Siegel, “The Equity Premium: Stock and Bond Returns Since 1802,” Financial Analysts Journal 48, 1 (January/February 1992): 28–46.

-----------------------
Case 16 The Boeing 7E7 173

Case 16 The Boeing 7E7 171

Case 16 The Boeing 7E7 169

Case 16 The Boeing 7E7 167

Case 16 The Boeing 7E7 175

Case 16 The Boeing 7E7 177

Suggestions for complementary cases on the cost of capital: “Teletech Corporation, 1996” (Case 15); “Nike Inc.” (Case 14) and “Coke vs. Pepsi” (UVA-F-1340)

Tip: The instructor can streamline the discussion considerably by dictating “base-case” assumptions for student use regarding cost of debt, risk-free rate, and equity market risk premium. Such direction could be offered in the assignment questions. Our preference is not to dictate very much, but rather, to let students explore alternatives. The balance of this note follows this path. However, many schools have a “house view” about the equity market risk premium. For this reason, the case does not cite an EMRP—the instructor is assumed to offer guidance here. At the date of the case, the four main estimates of EMRP were:
• 6.4% (geometric mean over T-bills)
• 4.7% (geometric mean over T-bonds)
• 8.4% (arithmetic mean over T-bills)
• 6.4% (arithmetic mean over T-bonds)
See the Appendix for definition and discussion of the geometric and arithmetic EMRPs. The calculation in Exhibit TN8 illustrates the use of the arithmetic mean over T-bills.

Question 1: Why is Boeing contemplating the launch of the 7E7 project? Is this a good time to do so?

Question 4: Let us examine the details of how to estimate the WACC. Let us go step-by-step. Where shall we get started?

Question 5. The weighted-average cost of capital is a simple formula. Yet it seems that reasonable people will disagree about the estimates. What are the points of contention in estimating the cost of capital? Please summarize the issues.

Question 7. Have we thought of everything? Is there anything else the board of directors should consider in assessing the financial appeal of this project? Why might the board vote “yes” on the 7E7 when the cost of capital estimate is greater than the IRR? Why might the board vote “no” if the cost of capital estimate is less than the IRR?

Question 8. What should the board do?

168 Case 16 The Boeing 7E7

--./WXYbc?’˜š¿ÀÈÉèéû[pic] w €
À
Â
2
7
¥
¦ ö ý
¦
ª
˜™ST[\abij89ÇÈàýC¶hY4yh]G5?hY4yh]G6?]?hv)Bh170 Case 16 The Boeing 7E7

172 Case 16 The Boeing 7E7

174 Case 16 The Boeing 7E7

176 Case 16 The Boeing 7E7

178 Case 16 The Boeing 7E7