Mgt 544 Investment Management

Case 2:     Enron Corporation

Now that Enron is being talked about everywhere, we want to look at (i) the investment valuation aspects of Enron (before Enron became a public scandal) and (ii) the financial risk aspects of Enron's structured-finance deals. That is, could one have found out months in advance that Enron was not a good investment? What should have been taken as indicators/factors that Enron was going to be in serious troubel? What was Enron's  risk profile before the discovery? How can we measure the risk of a structured financial transaction and that of a prepaid swap?

For articles on recent developments and background information about Enron, you can do a search using any internet search engine or the New York Times or Wall Street Journal. For this case, we want to focus on the following issues.

1. Pre-scandal valuation and investment analysis. On March 5 2001,  Fortune magazine published an article "Is Enron Overpriced?" . In this article, the author Bethany McLean contrasted Enron with  Duke Energy, Cisco, Goldman Sachs, and the S&P 500 index. For our analyses here, you will need to download historical financial data and earnings forecast data from Bloomberg. (See a related article by an analyst at Off Wall Street and theStreet.com.) Some of the data needed for this part are in this Excel file.

1.1.     Compare the (i) revenue growth, (ii) operating cashflow growth, (iii) earnings growth and (iv) profit margin, across  Enron, Duke Energy, Goldman Sachs, and Cisco. Use the annual data from 1995 to 2000. Make investment recommendations based on this comparison. (Please  ignore your knowledge of what happend after March 2001. Instead, the analysis should be as of March 2001). Discuss differences across these firms.

1.2.     Apply both the earnings discount model and the residual-income model (see Lee, Myers and Swaminathan paper) to value these four stocks as of January 1, 2001. For both models, use a 5-year valuation horizon, that is, discount each of the 5 years from 2001 to 2005 explicitly, and then estimate a terminal stock price as of the end of 2005 and discount that price back to the beginning of January 2001. For earnings-per-share (EPS) forecasts, use the consensus estimates for years 2001 and 2002. For the years 2003, 2004 and 2005, apply the consensus 5-year EPS growth estimate for each given firm and assume that this same growth rate estimate would apply to each of the three years for this firm. For the cost of equity capital, use the CAPM. For the end-of-2005 stock price (or, terminal value) estimate, follow the same method as given in Lee, Myers and Swaminathan paper (that is, use the same terminal value estimate for both the earnings discount and the residual-income models). For each stock's beta, use Yahoo Finance's company "Profile" page, or estimate your own betas using historical data. But, in any case, you should use only data available up to January 2001 to estimate the betas, EPS forecasts, costs of equity capital. Make investment recommendations based on these valuations.

1.3.     How would you modify your recommendations from part 1.2 if you take into account the 12-month "price momentum" factor? What other indicators or factors would have prevented you from buying or holding onto Enron back in early 2001? For historical price quotes for each stock, again you can rely on Yahoo Finance's Chart function.

2. On Sunday February 17, 2002, the New York Times published an article "Enron Had More Than One Way to Disguise Rapid Rise in Debt." It described various swaps that Enron used to effectively borrow loans without having them booked as debt. One of such transactions was done between Enron and CS First Boston in 2000.  In this prepaid swap, CS First Boston paid Enron $150 million up front on, say, December 31, 2000. This payment can be thought of as the present value of a two-year fixed-rate bond, so it could be viewed as one side of a typical swap deal (in which one party commits to paying a fixed-interest-rate loan and the other party to paying a floating-rate loan with the same principal bond value). In exchange, Enron would make  four semi-annual payments (the first on June 31, 2001 and the last on December 31, 2002), with the last payment including both the $150 million principal plus the last half-year interest. The exact details of the swap are not known yet. So, let's try to figure them out. One thing known about the contract is that the interest payment on the loan amount was linked to oil price changes.

2.1.  Assume that at the time of a semi-annual payment, the semi-annual interest rate equals some "fixed amount" (denoted by X) plus the rate of change in oil price over the recent 6 months. Let the rate of change in oil price observed over the recent 6 months be denoted by Y. Then, the semi-annual interest payment is given by $150 million times (X+Y). Of course, the four semi-annual interest payments in 2001 and 2002 would differ from each other in general. But, when CS First Boston and Enron had to decide the payment terms, they had to figure out a value for X (realizing that Y could not be known with certainty as of December 2000). Then, what value for X would be fair to both parties?  To make this determination, assume that the average oil-price change (i.e., expected oil-price change) per year is 6.2%, so the average oil-price change per 6 months is about 3.1%. Also, the discount rate to use for discounting the expected future payments is assumed to be the same as the cost of equity capital for Enron.

2.2.     The annual volatilty (standard deviation) of oil price changes is historically about 39.5%. Assume that oil price changes are normally distributed. Then, over any 6-month period, what is the value-at-risk (VAR) of this $150 million swap at the 95% confidence level? Who was bearing this risk of the swap?

3.     Off-balance sheet trusts were another type of structured-finance instrument that Enron used extensively. Marlin II is one such example. More details of the structure are given in a TheStreet.com article back in October 22, 2001, and a second article on October 23, 2001. You should read both of these articles to see the details. Marlin Water Trust II together with Osprey Trust (also known as Whitewing trust) were the two trusts that had to be repaid back by Enron  on November 28, 2001 and brought Enron to file for Chapter 11 on December 2.  Let's take Marlin II as an example, because Osprey was similar. (For a general, brief introduction to securitization, see a good article and other references.)
    Marlin II was created as a Special Purpose Entity that was used by Enron to issue $1 billion debt. The debt issue was backed by the water assets "given to" Marlin II by Enron. That is, Enron took these assets off its balance sheet and "sold" them to Marlin II, with a "sale" price of $1 billion. Backed by these assets, on July 1, 2001 Marlin II issued $1 billion debt maturing on July 1, 2003. In addition, the debt issue came with some guarantees and assurances by Enron, such that (i)  in case Marlin II would not have enough capital to pay for the debt obligations Enron would make up the difference, through issuing more Enron stock if necessary, and (ii) Enron would have to retire the debt for its full value of $1 billion, by issuing more Enron stock if necessary, if both of the following triggers would be satistfied at any time prior to the debt's maturity: (1) that Enron stock price go below $34.13 and (2) that Enron's credit rating fall below investment grade.

3.1.     Describe the "assets" and the "liabilities" of Marlin II.  How would you describe Enron's position in the Marlin II structure? Was Enron effectively the (unlimited-liability) residual claimant in the trust?

3.2.    Let's try to analyze the "assets" and "liabilities" of Marlin II as if they represented the "long" and "short" sides of a hedge-fund strategy (remember the duration-matching idea for fixed-income asset-liability management?). Then, was there a liquidity (or, illiquidity) mismatch between the long and the short side in this case? Was there an intrinsic difference in valuation difficulty between the long and the short side? (Generally, we think that marketed securities, especially those that are actively traded, have better liquidity and hence easier valuation than those assets that are not traded). To help put this in the right perspective, you should read the description of the conference call discussions on how much the water assets were worth in the second article on October 23, 2001.

3.3.     Let's assume that Enron was the effective residual claimant in Marlin II. Then, whatever risk remained in Marlin II that was not neutralized between the long and the short side would be 100% assumed by Enron. After taking out Enron's guarantees and assurances, we realize that the long side of Marlin II consisted of the water assets, and that the short side consisted of the $1 billion debt issued by Marlin II. How would you describe the degree of risk neutralization between the two sides? Since the structure used in Marlin II is a very typical asset-based securitization structure in the industry, what does this long-short risk-neutralization analysis tell you about the risks of such securitization efforts?
    To quantify your analysis here, assume that the value of the water assets was perfectly correlated with Enron's stock price, but had 60% of the total volatility of the latter. On the other hand, the 2-year debt issued by Marlin II had an expected annual return of 6%, an annual volatility of 9% and a correlation of 0.2 with Enron's stock price (consistent with the historical experience of high-quality corporate bonds). First, use Enron's monthly stock returns for 1998, 1999 and 2000 to estimate Enron's stock return volatility. Next, determine the remaining risk (i.e., annual volatility) of the Marlin II long-short positions. That risk is the un-neutralized risk of the trust assumed by the residual claimant.

=============================
For data-related questions, contact  Zheng Yue