Testimony of Charles L. Hill Director of Research Thomson Financial / First Call before the joint session of the House Subcommittee on Capital Markets, Insurance, and Government Sponsored Enterprises and the House Subcommittee on Oversight and Investigations 12 October 2001 BROKER ANALYST'S ANALYSIS OF ENRON Prologue I welcome the opportunity to again testify in front of the House Financial Services Committee. I believe this committee has been addressing substantive issues that are important not only to the future health of the investment community, but important to the general public's perception of and confidence in the overall capitalist system. The excesses associated with Enron that led to its bankruptcy are more far reaching than just their impact on Enron. There is plenty of blame to go around in the mistakes made in the Enron situation. I am here today to focus on the role of the broker analysts in this debacle. In my previous testimony before this committee, I did not tread lightly on what I thought were some serious problems in analyst behavior that needed to be remedied. I am here this morning, however, to say that the analysts to some degree were more victims rather than culprits in the Enron situation. Not that they were without blame, particularly in the late stages of the Enron collapse, but they were not the underlying cause of the excessive rise in Enron's stock that later proved to be irrational. The performance of the analysts should be judged on two fronts. The first is their analysis of Enron's fundamentals, particularly in regard to earnings. The second is their valuation assessment and recommendations of Enron stock. Analysis of Enron Fundamentals by Broker Analysts The thing that stands out most visibly about the analyst's analyses of Enron, is that over the three years up to October 2001, their estimates at the beginning of each year for that year had minimal changes. The few changes that did occur were always upward and usually followed the guidance given by the company when they reported quarterly earnings. The narrowness of the spread of estimates among analysts was remarkable, especially for an energy company. The coefficient of variance for Enron estimates was consistently below the average for the S&P500 during the same period. This pattern is highly suggestive that the analysts were being spoon fed as to what Enron expected earnings to be. The analysts might have been willing to accept company guidance, be it overt or inferred, as long as the company kept meeting expectations each quarter. Since at least the beginning of 1998, Enron has met or exceeded analyst estimates every quarter. One reason that analysts may have been more willing than normal to accept company guidance for Enron was that it was becoming increasingly difficult to understand how Enron was achieving its revenue growth and profitability. Extensive use of derivatives, particularly when the company is using mark-to-market accounting is extremely difficult in the best of situations. We now know that a big additional reason for the difficulties in analyzing Enron's financials was that there were significant parts of Enron's business that were hidden from the balance sheet. Often the way out for analysts when faced with difficult to analyze situations like Enron is to drop coverage. Why take the risk when there are plenty of companies that are transparent enough to do meaningful analysis with confidence? The problem with dropping Enron was that it had become the giant in the industry. If you were an analyst covering that industry, you essentially had to cover Enron. That was further reinforced if your firm was one of Enron's investment bankers or investment banker wannabe. The real problem though was having sufficient information about the off balance sheet items. Whether the accounting for each of these items was within FASB rules or not is not yet clear, although the announced restatement of prior periods earnings is a strong signal that at least not all was kosher. But what is clear is that Enron was not providing what could even be considered minimum transparency in its financials and that the analysts did not have all the tools necessary to make a reasonable analysis. Valuation of Enron Stock by Broker Analysts In evaluating analyst performance on recommending Enron stock, one first has to understand how the brokerage community's recommendation really works. As I have testified before to this committee, the investor needs a two level decoder. The first level of the decoder gets all the brokers on a common recommendation scale. The most common scale is a five tiered one, where the top category is a strong buy, the second is a buy, the third is hold, the fourth is sell, and the fifth is strong sell. Most brokers have a five tier scale, some have a four tier one, and a few have a three tier scale. In addition, many have very different terminology. The term "buy" may be the term used for the top category at some brokers, or for the second best category at many brokers, or, in at least one case, for the middle category. There are more than a dozen different terms used for each of the top three categories, and almost as many for the bottom two. Unfortunately, getting all the firms on a common scale is not the end of the decoding. Analysts are overly biased on the positive side in their recommendations. The typical distribution is about 33% of all recommendations are in the top or strong buy category, about 33% in the second or buy category, about 33% in the middle or hold category, and only about 1% in the remaining sell and strong sell categories combined. If the recommendations are put in numeric terms where 1 is a strong buy (or whatever the broker's term is for their top category), 2 is a buy, 3 is a hold, 4 is a sell, and 5 is a strong sell. Using this numerical scale, consensus recommendations can be calculated for each company. Most of the time the average consensus recommendation for either the companies in the S&P500, or for the roughly 5000 companies that analysts cover, is a 2.1. Occasionally, the average may be a 2.0 or a 2.2. Therefore, the second level of the decoder would move the recommendations into three more meaningful categories. Those in the 1 or strong buy category would really be saying buy, at least in relative terms. Those in the 2 or buy category would really be saying they were neutral on the stock, and those in the 3 or hold, the 4 or sell, and the 5 or strong sell categories all would be saying sell the stock. For Enron, the consensus recommendation, as shown on the accompanying graph, was about a 1.5 from May 2000 until the end of September 2001. Even if we had our decoder to compensate for analyst optimism, it is clear that the analyst covering Enron were very positive with their recommendations. But during that same period, the analysts had similar or higher consensus recommendations on competitors like Calpine and Dynegy. While the consensus recommendation for Enron was much better than the average for S&P500 companies, there enthusiasm was not limited to Enron. In early October 2001, the consensus recommendation spiked up from a 1.5 to a 1.3 as several analysts raised their recommendations ahead of Enron reporting its 3Q01 earnings on 16 October. On the day of the earnings announcement one analyst raised their recommendation, pushing the consensus to a remarkable 1.2. But as the Enron story began to unravel over the next few days, the recommendation downgrades exploded, plus six of the seventeen analysts dropped coverage. Conclusions In these kind of situations, it is easy to point a finger at the analysts for mistakes made. In my prior testimony, and in other forums, I have taken the analysts to task for not performing to an acceptable standard in certain situations. While the analysts are certainly not without blame on Enron, they are not the real culprits in this situation. I am not an expert in doing the actual accounting at a company, or in auditing a company's accounting, but having been an analyst for 22 years, as well as closely observing analyst behavior at First Call for the last ten, I can say without reservation that this was a situation where either the company or its auditors or both were at fault in not providing investors, especially including the analysts, with the tools necessary to understand Enron's business. Whether the letter of the accounting rules were met or not, it is patently obvious that the spirit of the rules was violated in that Enron's financial statements did not fairly convey enough information for investors to reasonably analyze the company's operations. In that climate, it is hard to be too critical of the analysts' optimism. Enron had a long history of showing consistent and substantive earnings growth. If it had been up to me if I was in that situation, I would have dropped coverage long before October 2001. The financial reports and details of operations had become more and more inscrutable well before then. But as I mentioned earlier, most, if not all, analysts did not have that operation. All things considered, they probably did as well as could be expected until October 2001, although in hindsight it is easy to say that they could have at least tempered their bullish recommendations to some degree. However, once the issues of the off balance sheet items became an unexplained issue on the 16 October 2001 conference call on 3Q01 results, it does seem that the analysts could have moved quicker to either suspend their recommendation or dramatically drop the level of their recommendation. The unexplained $1.2 billion balance sheet writedown was not a caution flag, it was a red flag. But Enron is not the situation on which to challenge analyst performance. There are far more significant situations were analyst conflicts and performance are at issue. The lessons to be learned here is how to insure that company's and their auditors can be relied on to openly provide the necessary tools for investors to meaningfully analyze the company's business. 100 90 80 70 60 50 40 30 20 10 0 20 21 23 1 18 19 /0 3 / 15 2 16 17 22 1 1 13 14 /0 3 / 1 11 1 10 1 9 /0 3 / 0 8 ec 1 01 7 / ice 3 / 9 cons r pr 01 / 3 / 8 01 / 3 / 7 01 / 3 / 6 01 / 3 / 5 on 5 01 / 3 / 4 01 / 3 / rong sell = 3 01 / 3 / 3, St 2 01 / 3 / 10/0 1, Hold = 3 / 2 1 0 /0 6 3 / 1 1 0 /0 rong buy = 3 / 0 1 Enron Consensus Recommendati St 00 / 3 / 9 5 00 / 3 / 8 00 / 3 / 7 00 / 3 / 6 00 4 / 3 / 3 5 00 / 3 / 4 2 00 / 3 / 3 1 00 / 3 / 2 0 0 0 0 0 00 / 1. 2. 3. 4. 5. 3 / 1 2.00 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 4Q01 3Q01 2Q01 tr EPS 1Q01 ice 4Q00 Pr Trlg 4 Q 3Q00 2Q00 1Q00 arnings e 4Q99 Qtr 3Q99 2Q99 ailing 4 1Q99 nd Tr 4Q98 e aic 3Q98 2Q98 on - Pr 1Q98 Enr 4Q97 3Q97 2Q97 1Q97 4Q96 3Q96 2Q96 1Q96 10.00 100.00 1999 2000 2001 2002 FY FY FY FY 2002 1 FY 0 / 1 2001 / 1 1 FY 1 /0 1 / 9 1 /0 1 / 7 1 /0 1 / 2002 5 2000 1 FY /0 1 / 3 1 /0 1 / 1 or FY 1999 - 0 0 / 1 / ory f 1 1 st 0 /0 1 / 9 e Hi 0 /0 1 imat / 7 1999 0 FY /0 1 / 5 ngs Est 0 /0 1 / 3 0 /0 1 / 1 Enron Earni 9 9 / 1 / 1 19/91/9 9 /9 1 / 7 9 /9 1 / 5 9 /9 1 / 3 9 50 00 50 00 50 00 /9 1 / 2. 2. 1. 1. 0. 0. 1