USEC INC. Teaching Note

Transcription

1 Nov. 5, 2008 USEC INC. Teaching Note This case presents the student with the challenges of formulating a discounted cashflow (DCF) analysis for a strategically important capital-investment decision. Analytically the problem is representative of most corporate-investment decisions, but it is particularly interesting because of the massive size of the American Centrifuge Project (ACP) and the potential of the project to significantly affect the stock price. Students must determine the relevant cash flows, paying close attention to the treatment of input costs, selling prices, the timing of investment outlays, depreciation, and inflation. An important input is the appropriate cost of uranium, which some students argue should be included at book value, while others argue that market value should be used. Although the primary objective of the case is to focus on the estimation of cash flows, students are provided with a straightforward set of inputs to estimate USEC s weighted-average cost of capital. It is an appropriate case for students who are learning or need a refresher on DCF analysis. Because of the basic issues covered, the case works well with undergraduate, MBA, and executive-education audiences. The case also affords the opportunity to explore a variety of issues related to capital-investment analysis, including relevant costs, incremental analysis, cost of capital, and sensitivity analysis. The case is an excellent example of estimating the value of a firm as the value of assets in place plus the net present value of future growth opportunities. Suggested Study Questions for Students 1. Why is USEC pursuing the ACP? 2. What is the net present value (NPV) of the ACP currently implied by the stock market? 3. What is USEC s weighted-average cost of capital (WACC) in July 2006? Are you comfortable with using this discount rate for the ACP? This teaching note was prepared by Benjamin Mackovjak (MBA 07) under the supervision of Kenneth Eades, Professor of Business Administration. Copyright 2007 by the University of Virginia Darden School Foundation, Charlottesville, VA. All rights reserved. To order copies, send an to sales@dardenbusinesspublishing.com. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means electronic, mechanical, photocopying, recording, or otherwise without the permission of the Darden School Foundation.

2 -2-4. Using the financial data in the case, what is your estimate of the American Centrifuge Project s NPV? What are the key value drivers in your estimate? Use a sensitivity analysis to illustrate how much the NPV varies for reasonable changes of the key driver inputs. 5. Is Mackovjak correct to ignore that the market value of USEC s inventory of uranium exceeds its book value? 6. Based on your analysis, would you recommend that Rivanna Capital take a long, short, or no position in USEC s stock? Teaching-Plan Outline The following teaching plan is for an 85-minute class period. (10 min.) Discuss USEC s business model and current operating environment. (10 min.) Discuss the American Centrifuge Project, the expenditures required, and the benefits presented by the project. (10 min.) Estimate USEC s WACC and critique the assumptions used in the estimation process. (40 min.) Present and critique cash-flow estimates. Discuss NPV and sensitivity analysis of value drivers. (10 min.) Discuss the market value of the inventory versus book value. The stock may still be a good buy even if the NPV of the ACP is negative. (5 min.) Epilogue and concluding discussion. USEC s Business Model USEC was a unique company with both economic and political origins, the leading producer and supplier of enriched uranium for commercial nuclear-power plants, and the executive agent for the Megatons to Megawatts program that recycled nuclear warheads into electricity. As a publicly traded company, USEC faced the same scrutiny regarding its financial performance as any other company, but USEC s financial performance was connected with parts of the Megatons to Megawatts agreement, which kept its cost of goods sold (COGS) low, relative to buying uranium on the open market. Fundamentally, the keys to USEC s success were the supply of uranium, execution of the enrichment process, and the demand for SWUs. The most controllable factor was the enrichment process where USEC could significantly reduce the cost by investing in the ACP. Some students will recognize that USEC had attempted to manage its relationship with the cost of electricity by the use of a power-supply contract. It was the lapse of this contract that

3 -3- recently had put pressure on USEC s margins, making it all the more important for the company to find a way to reduce its operating costs to remain competitive. The instructor may help students think more critically about USEC and its stock price by asking whether it was truly a public company or should be considered as an operational branch of the U.S. government: The answer lies somewhere in the middle. USEC was a publicly traded company and as such, presumably could be taken over by an outside management group. At the same time, it seemed unlikely that the United States would allow a foreign company to gain control of the company because of security issues connected with nuclear energy. To the extent that the company has a limited number of potential buyers, the possible upside for an investor would be limited, relative to most other publicly traded companies. Still, it was highly unlikely that the United States would allow USEC to fail. In addition to enriching uranium for use in power generation, USEC also enriched uranium for sale to the Department of Defense for use in nonproliferation of nuclear weapons. Therefore, given the strategic importance to national defense, USEC probably enjoyed a certain downside protection to its stock price as well. USEC s debt rating of B suggests that the company had not benefited greatly from its strategic importance, as the market apparently viewed it as having a significant risk of default, which would not be the case if USEC were viewed as being backed by the U.S. government. The American Centrifuge Project The cost savings of the ACP was primarily associated with reduction of electricity required for the enrichment process. Reducing the power consumption by 95% would contribute to an overall savings of 50%, relative to the existing gaseous-diffusion process used currently by USEC at its Paducah facility. Viewed in isolation, the strategy of investing in the ACP was sensible because it would have placed USEC as the cost leader in the industry. The problem for students to realize is that the savings required an extremely large investment, and the benefits occurred over a long period. I like to challenge the class by asking, USEC has a $900 million market and is spending $1.7 billion on the ACP, so how could it be possible that this investment will not increase the value of the company? The trick to the question is that the investment will certainly increase the value of the enterprise, but there is no guarantee that it will increase the stock price. The impact upon the stock price may only be determined by conducting a discounted-cash-flow analysis. This is the same approach used to evaluate any capitalinvestment project. The difference is that the ACP happened to be an extremely large undertaking, making the analysis all the more important for USEC s shareholders as well as for Ben Mackovjak, who was trying to estimate the intrinsic value of the company as an investment. Despite the size of the ACP and despite the long investment horizon, the basic principles of discounted cash flow still hold. Students need to estimate the appropriate cost of

4 -4- capital and use that rate to discount the future free cash flows produced by the project. As a starting place, I ask students to estimate USEC s WACC and have them decide whether it is the appropriate rate to use for the ACP. USEC s WACC Exhibit TN1 displays an estimate of USEC s WACC. The cost of debt is given in the case as 9.04%, the yield to maturity of USEC s outstanding long-term debt. Case Exhibit 5 also gives the 30-year Treasury yield as 5.09% and USEC s beta as 1.3, which together with a 6.0% market-risk premium give a cost of equity estimate as 12.9%, using the capital-asset pricing model. Using current market values of debt and equity give market weights of 32% and 68%, respectively, and a WACC estimate of 10.5%. Some students will question whether USEC s current cost of capital is appropriate for the ACP. The size of the investment and the risks of the new technology often prompt students to add a risk premium to USEC s WACC. Other students will argue that it is only systematic risks that are priced in an efficient capital market, so USEC shareholders would not demand a higher return rather than for risks that are largely independent of the market and therefore mostly diversifiable. Students often struggle with whether risks should be incorporated in the discount rate or in the cash flows; however, it is important to avoid the mistake of increasing the discount rate for nonsystematic risks. It is true that a new project carries a variety of uncertainties, but most of them are incorporated in the estimation of the cash flows. The key to a good cash-flow forecast is to be aware that DCF requires that expected cash flows be discounted, as opposed to the best-case cash flows. An expected cash flow incorporates probabilities for downside and upside scenarios and, in particular, for the ACP and explicitly considers the chance that the ACP technology never proves viable. This is equivalent to scaling future cash flows according to the chance of success (e.g., multiply the operating cash flows by 0.9 if the probability of failure is 0.1). Increasing the discount rate is reserved for those situations when the analyst has market data supporting a higher discount rate, such as the cost of capital for a pure play company that approximates the risk as a new venture. I like to let this discussion run for several minutes to see if the class arrives at a resolution between cash-flow adjustments and adjustments to the discount rate. If a consensus does not arise, I move forward with the WACC estimate with the promise of returning to the issue later in the class as part of the sensitivity analysis. DCF Analysis Most of class time is spent going over the estimation of the free cash flows for the DCF analysis. Exhibit TN2 contains a summary of the free cash-flow estimates along with the key

5 -5- assumptions. The ACP represents USEC s bid to become the low-cost producer in a highly competitive, commoditized market. In addition to the cost reductions, the cash flows also reveal that the ACP investment will be a revenue-increasing undertaking. The instructor may want to start by asking the class for ideas about what relevant cash flows to consider for the analysis. After creating a format similar to that shown in Exhibit TN2, the instructor may ask individual students for an estimate of each line item. I suggest doing the first couple of operating years plus the final year to save time. As the numbers are filled in, students will discuss their points of disagreement with what is being recorded, making for a lively discussion of the assumptions used for the analysis. The case works well for both novice and experienced students in that the instructor may choose to devote time to the fundamentals of DCF or to the subtleties for the more experienced students. Regardless of their experience level, there are some interesting issues that invite discussion from students. 1. Is the $100 million a sunk cost? An interesting issue for the class to consider is how to handle the $100 million that has already been spent on the ACP. Most students will argue that the $100 million is a sunk cost and should not be considered for an NPV calculation. While students generally accept the sunk-cost argument, some will argue that to ignore this amount will make the project appear more valuable than it should. This could make for an intensive debate and a good learning experience for students of all levels. If no part of the $100 million can be retrieved, then it is a sunk cost and should be omitted from the analysis. What could be confusing for students is that the $100 million investment will be captured as part of the financing calculations, either as part of the debt financing or as equity in the form of the number of shares outstanding. Thus, if we were to include the $100 million in the estimate of NPV for the project as a negative cash flow, it would be double counted in the overall analysis of the stock price. Because the case does not explicitly explain how a DCF is used to compute equity value, the instructor may want to take the opportunity to present the valuation framework shown in Figure 1 and ask the class where the $100 million would show up. Figure 1. Valuation framework. Equity Valuation: D = market value of interest-bearing debt + E = market value equity value V = enterprise value = PV (free cash flows) E = V D Price per share = E/N N = Number of shares outstanding

6 -6- As USEC raises funds for the ACP, either the debt outstanding will increase or the shares outstanding will increase. 1 Either way, the value per share is reduced by the amount of the investment. In fact, over the ensuing six years, more and more of the investment dollars will be spent, and debt and equity outstanding will increase. Correspondingly, the NPV of the project will rise as more and more of the investment dollars become sunk costs for the NPV calculation. The net effect, however, will need to have included all the costs and benefits of the ACP in the equity valuation so that the value per share will compute correctly. 2. What should be the price of uranium? The case states that the MTM program allowed USEC to buy either raw uranium or the equivalent of 5 million pounds of uranium. Either way, USEC paid the equivalent of $20/lb. of raw uranium and recorded it as $20/lb. in its inventory account. In truth, almost all of the uranium purchased was enriched uranium that had been removed from Russian missiles. This uranium was simply tested for SWU content and then shipped on to a fabricator to be turned into fuel rods for nuclear-power plants. The case was written so students will conclude that USEC had stockpiled a large inventory of raw uranium that was being used as an input for the enrichment process. And while not completely accurate, it serves the pedagogical objective of prompting a very interesting debate regarding what to use for the cost of uranium in the analysis. Many students will argue that USEC should use $20/lb. in the analysis because it represents the actual cost of uranium for the company. Others will argue that the market price of $43/lb. is more appropriate because this avoids enhancing the value of the ACP with a benefit incurred through the MTM program. This debate is valuable for the students as it emphasizes the importance of relevant costs to a cash-flow analysis. Students may want to use $20/lb. to map out the inventory levels over the ensuing years to determine when USEC will run out of the cheap inventory. They will assume that for the remaining seven years of the MTM, USEC will continue to buy 5 million pounds of uranium per year to stockpile a huge inventory until the ACP comes on line, and the production level rises to draw down the inventory. This effort reflects exactly what accountants must do to value correctly the COGS and the uranium inventory. The accounting treatment, however, does not necessarily capture the true economic impact of using the uranium, which is more appropriate for answering the question about how much value the ACP adds to USEC. The NPV requires a comparison of the ACP cash flows with the cash flows that would have occurred without the ACP. These incremental cash flows include a cost-reduction component as well as a revenue-increase component, which are offset by the negative cash flows of $1.6 billion of the investment. Some students will recognize that the cost of uranium is common to both cash flows, and that for the most part, it becomes irrelevant to the incremental cash flows whether one uses market values or book values. The only incremental impact upon the incremental cash flows is the timing of when the $20/lb. uranium is used. 1 This assumes that USEC does not have excess cash to spend in lieu of issuing shares or new debt. Because cash is negative debt, the net effect of spending cash is to increase the net debt measured as debt less cash.

7 -7- Because the ACP has a higher production capability, the MTM uranium supply will be used sooner than under the current production capacity of the Paducah plant. This changes the time value of when the savings of the uranium is realized. The timing difference, however, occurs so far in the future that it has a minimum impact upon the NPV, making the choice between book and market value a minor issue for the analysis. Other students will recognize that by using $20 we are taking value being realized from the Megatons to Megawatts program and attributing it to the ACP. If we use the market price, the resulting NPV will reflect the value of the ACP in a competitive market, which is a better measure of how the market should be viewing the incremental value of the ACP. This assumes that USEC would either use the excess uranium or sell it at its market price in the open market. USEC would not want to dump a large amount of uranium on the market at one time, but it should be able to sell it gradually over a reasonable period without significantly depressing the market price. Admittedly, if we assume that USEC cannot sell the uranium in the open market and that the ACP is the only vehicle for using the uranium faster, then the project should be credited for the time value gained by realizing the uranium savings sooner. Because the case states that USEC did sell its inventory in the market, the DCF presented in this note used the market value of uranium to reflect that opportunity. 3. How can we handle construction risk and operation risk? The base-case cash flows assume that construction will be completed on time and that the ACP will operate successfully. Both of these factors should be considered key value drivers as they have significant impact on the NPV. The probability of successfully completing and operating the ACP may be included in the analysis by scaling the operating cash flows (except depreciation) by the probability of success. If Mackovjak feels that the project has a 90% chance of becoming operational, then the free cash flows after 2010 should be scaled by 0.9. Similarly, we could model construction risk by scaling the construction costs, which are the negative cash flows for If a cost overrun of 10% is a reasonable consideration, then the construction costs could be scaled by a factor of 1.1 to reflect this scenario. If Mackovjak feels that a year might be added to the construction schedule, then he could simply postpone the inflows to begin in 2012, rather than 2011, and reduce the NPV by the time value of delaying one year. Each of these factors may compute NPV values as part of a sensitivity analysis. Both the base case in Exhibit TN2 and the sensitivity analysis shown in Exhibit TN3 suggest that the ACP was likely to be a negative NPV project but not negative enough to warrant Rivanna taking a short position in the stock.

8 -8- Uranium Inventory Exhibit TN4 is identical to case Exhibit 3, which calculates a net working-capital liquidation value on a book- and market-value basis. The case states that the inventory includes the equivalent of 29 million pounds of uranium that was recorded at the contracted price of $20/lb. Thus, the primary difference between the book- and market-value calculations is the $580 million (29 $20) of MTM inventory and its market value equal to $1,247 (29 $43). At the market value of $43/lb., the MTM inventory plus the other net working capital would more than cover USEC s debt (long-term and pension liability). In fact, such liquidation would net the shareholders $2.50 per share, a 23% premium over the current stock price of $ This is a startling statement about USEC and its management. At a minimum, it suggests that USEC s business model is less valuable than the value of the MTM. Students should realize that in an efficient market, such an anomaly would not exist. The question, however, lies in how the shareholders could unlock this value. If USEC could be bought out by a private equity fund, for example, the market would likely have adjusted the stock price upward to reflect the potential of this liquidation value. Yet most students will argue that the company s strategic value to the United States makes it unlikely that the firm would be taken over and liquidated. In this sense, we may not be surprised that the market attaches no value to a liquidation value that cannot be realized. Epilogue This case demonstrates the principle of value additivity whereby the value of the firm equals the value of the assets in place plus the value of its growth opportunities. For USEC, ACP represented the primary growth opportunity so the value of USEC stock equaled the value of ongoing operations plus the NPV of the ACP. Mackovjak s analysis closely paralleled that of the one presented in this note in that he found the value for the ACP to be approximately zero, and he recommended that Rivanna should take neither a long nor a short position in USEC. After July 2006, USEC s stock price continued to vary. By May 2007, the price had more than doubled to $23, making Mackovjak a bit nervous about his recommendation. By the end of 2007, however, the stock had fallen below $8 (Exhibit TN5). The downturn was prompted in part by analyst reports, which had begun to express concern about the value of the ACP and the many risks associated with such a large project.

9 -9- Exhibit TN1 USEC INC. Estimation of Weighted Average Cost of Capital Risk-free rate (30 year) 5.09% Market risk premium (assumed) 6.00% Beta (Exhibit 5) 1.3 Cost of equity 12.9% Cost of debt (Exhibit 5) 9.04% Tax rate (approximately 2005 Rate) 40% After-tax cost of debt 5.4% Stock price $10.80 Number of shares outstanding (millions) 86.1 Market value of equity (millions) $929.9 Weight of equity to capital 68% Book value of debt (millions) $475.0 Market value of debt (millions) $435.2 Weight of debt to capital 32% Weighted Average Cost of Capital 10.5%

10 -10- Exhibit TN2 USEC INC. ACP Base-Case Free Cash Flows and NPV SUMMARY of ASSUMPTIONS and RESULTS American Centrifuge Project (ACP) Gaseous Diffusion Process Price of Separate Work Units (SWU) $127 SWU Sold Currently 3.5 millions Royalties for use of Centrifuge Technology 1% Lease Expense $8.0 millions SWU sold in year 2011 (millions) 2.5 Depreciation $30.0 millions SWU increment for years Maintenance Capex $30.0 millions Market Price of Uranium per SWU $43 Inflation 3.0% Cost of Uranium per Pound (Book Value) $20 WACC 10.5% Uranium Enrichment Cost per SWU $42 Tax Rate 40% Depreciable Years 15 CAPEX $30.0 millions Target Investment (millions) $1,700 Capital Invested (millions) ($1,700) Cost Overrun 0% ACP Incremental NPV/IRR Probability Cost Overrun 100% IRR = 9.8% Reduction in Enrichment Cost 50% NPV = ($67.7) Reduction in CF (probability of success) 100% NPV/Share = ($0.79) Net Working Capital as % of Sales 5.0% NPV/Price = -7.3% SG&A 4.0%

11 -11- Exhibit TN2 (continued) American Centrifuge Project (ACP) Year > Volume of SWU sold Gross Revenue $368.1 $682.4 $1,405.4 $1,447.5 Net Revenue $364.4 $675.6 $1,391.3 $1,433.0 Cost of Goods Sold : Cost of Uranium Cost of Enrichment Depreciation Gross Margin SG&A EBIT NOPAT Depreciation Net Working Capital Net Working Capital Change 6.6 (15.7) (2.0) 70.3 Cash Flow Operations Cash Flow prob of success Capital Investment ($185) ($300) ($350) ($350) ($415) Total Cash Flow ($185) ($300) ($350) ($350) ($415) $150 $212 $419 $503

12 -12- Exhibit TN2 (continued) Air-Diffusion Process Operating cash flows are identical through 2010 Year > Volume of SWU sold Revenue $444.5 $457.8 $471.6 $485.7 $500.3 $368.1 $530.8 $756.7 $779.4 Cost of goods sold: Cost of Uranium Cost of enrichment Depreciation Gross margin Operating lease expense SG&A EBIT NOPAT Depreciation Net working capital Net-working-capital change (0.7) (0.7) (0.7) (0.7) 6.6 (8.1) (1.1) 37.8 Capital expenditure Cash-Flow Operations Combined Cash Flows Air diffusion process (current process) $44 $57 $92 $134 American Centrifuge project (ACP) ($185) ($300) ($350) ($350) ($415) $150 $212 $419 $503 Incremental Cash Flow ($185) ($300) ($350) ($350) ($415) $106 $ > $327 $369 NPV = ($67.7) IRR = 9.8%

13 -13- Exhibit TN3 USEC INC. Sensitivity Analysis of ACP NPV NPV Base case ($67.7) Uranium = $20/lb * ($42.0) Inflation = 0% ($345.2) WACC = 11.5 % ($159.5) Cost overrun = 10% ($174.6) Probability of tech failure = 10% ($224.6) * Assumes ACP can use $20/lb through 2017, whereas current production can use it through 2022.

14 -14- Exhibit TN4 USEC INC. Net Working-Capital Liquidation Value ($ millions) Book Market Cash and short-term investments Accounts receivable Trade, net Megatons to megawatts inventory * ,247.0 Other inventory (SWU, uranium, etc.) Total current assets $1,495.2 $2,162.2 Accounts payable Customer advances Other payables Total current liabilities $428.0 $428.0 Net working capital $1,067.2 $1,734.2 Less debt ($475.0) ($437.8) Less pension underfunding ($153.9) ($153.9) Liquidation value $438.3 $1,142.5 Liquidation value/share $5.1 $13.3 Premium over current stock price ($10.8) (53%) 23% * Market value of uranium computed as 29 million pounds $43/lb. = $1,247 million

15 -15- Exhibit TN5 USEC INC. Stock-Price Performance