The energy investment-banking team at Hibernia Southcoast Capital in New Orleans recently analyzed the effect of return on capital employed (ROCE) on a public company's stock price performance. Developed from March 2003 to February 2004, the study analyzes financial data from 2001 to 2003 for 55 public U.S. companies. Southcoast chose to include data from three energy industry sectors: 21 exploration and production companies, 21 oil-service and 13 contract drillers. The team initiated the study after noticing that ROCE seemed to be gaining importance as a metric to gauge stock performance. Several energy executives discussed ROCE at various industry conferences, offering their opinions on the effect of ROCE on their companies' stock performance. In May 2000, Oil and Gas Investor reported that, during a popular energy industry conference, ROCE "was the new mantra chanted by nearly every executive." Southcoast's investment-banking group applied linear regression analysis to a broad sample of energy companies to determine more accurately how effectively ROCE drives stock performance. Given such extensive industry acceptance, ROCE would be expected to have a significant effect on each of the three industry sectors analyzed; however, that proved not to be the case. Contrary to conventional wisdom, Southcoast found that E&P companies experience almost no correlation between ROCE and stock performance. In fact, net asset value (NAV) is a far better predictor of stock price performance. One probable reason for this high correlation is that investors reward E&P companies for finding and replacing reserves more than for prudent management of their capital structure. Therefore, an E&P company's stock price will perform better if managers invest debt and equity capital on initiatives focused on increasing NAV while controlling finding and development costs. As expected, Southcoast found that oilfield-service and contract-drilling companies do experience a high correlation between ROCE and stock performance. The strong relationship exists because investors reward companies that effectively use debt and equity for investment in their property, plant and equipment. This investment is used to generate net income that provides an acceptable rate of return. The positive correlation between earnings growth and stock performance has long been understood, but earnings growth alone does not take into account the balance sheet. The high correlation with ROCE for oilfield-service companies and drillers indicates that a company must grow earnings at a faster rate than it increases its level of capital, thereby increasing ROCE and ultimately improving the company's stock performance. Calculating ROCE For purposes of the study, ROCE is defined as net income before extraordinary charges, divided by the sum of long and short-term debt plus stockholders' equity. "Southcoast believes ROCE is one of the better analytical tools for understanding a company's performance in certain industry sectors, compared with other measures such as return on assets, because ROCE illustrates the financial performance of the company relative to its capital structure," explains Stan Ellington, managing director of investment banking at Hibernia Southcoast Capital. "Return on assets includes a number of inconsistencies between the numerator and the denominator. For example, long- and short-term debt is not included in the total asset denominator, while the interest payments for this debt are deducted from the numerator." There are two keys in applying the ROCE formula in such a thorough analysis: consistency in applying the formula across all companies and sectors, and resistance to making adjustments or excuses for less than desirable results. Southcoast showed its analysis to more than 25 energy industry executives as the study unfolded. Companies that did not perform well relative to their peers dismissed the study, saying it did not apply to their business model, or it was done over an improper time frame. Other companies thought that there were potential limitations in the ROCE formula that were the reasons for their poor performance. On the other hand, executives whose companies performed well believed the analysis confirmed their diligence in understanding their business drivers, be it NAV growth or maintaining an acceptable ROCE. Lloyd Carr, head coach of the University of Michigan football program in Ann Arbor, makes an analogy for this point: "Developing a high-performing organization involves making no excuses, acknowledging the issues and developing a strategy to resolve them. On the football field and in business, regardless of bad calls or missed opportunities, the end result is the final score." E&P results The 21 independent E&P companies Southcoast analyzed have market capitalizations ranging from $100 million to $12 billion. The analysis showed virtually no correlation (R2=0.02) between ROCE and stock performance. However, the study team found a strong correlation (R2=0.76) between an increase in NAV of proven reserves and stock performance, indicating that E&P companies are indeed valued on growth of reserves, not solely on the past returns on their invested capital. The results taken together suggest that investors should not necessarily focus on ROCE in the E&P industry, but look for companies that use debt and equity to effectively increase NAV. One top performer has been Fort Worth-based XTO Energy. From January 2001 to February 2004, XTO successfully grew NAV 40% while simultaneously experiencing a 126% increase in its stock price as defined in this analysis. Louis Baldwin, XTO Energy chief financial officer, says, "XTO understands both the benefits and limitations of ROCE as a financial measure. The strong correlation between NAV and stock performance demonstrated in this study is an indication of the value that investors put on our reserve growth." Quicksilver Resources also has done well. It has reported an ROCE of only 5%, but its stock price increased 169%-due to a 116% increase in NAV. By focusing on NAV growth rather than ROCE, Quicksilver exhibited the second-best stock performance among the 21 independents. Unfortunately, three other examples confirm this relationship between NAV and stock price, but on the negative side of the ledger. Royal Dutch/Shell, although not part of the study group, announced a 20% reduction in proved reserves. El Paso announced a 41% reduction and Forest Oil, a 35% reduction. These companies also were not studied. These events wiped out almost $10 billion or 8% of their collective market capitalization within days of the announcements. Oilfield services For the 21 oil-service companies that were analyzed with market capitalizations ranging from $500 million to $36 billion, Southcoast found a strong correlation (R2=0.54) between ROCE and stock performance. During the studied period, FMC Technologies exhibited ROCE of 12% while its stock appreciated 64%. The Houston-based firm has made ROCE a key metric for its managers. "FMC firmly believes that we need to both grow our earnings and maintain a high ROCE," says Joe Netherland, FMC Technologies chief executive officer. "Other analysis has shown that growing earnings was critical. This study confirms that ROCE, which considers a company's capital structure, also helps to drive shareholder value." Contract drillers For the 13 contract drillers with market capitalizations ranging from $380 million to $9.5 billion, that were analyzed, a strong correlation (R2=0.66) was apparent between ROCE and stock performance. The best-performing driller, based on ROCE, is Tulsa-based Unit Corp., which is a bit of an anomaly because the company has both a contract-drilling and E&P business that, when combined, exhibited an ROCE of 12% and a corresponding 41% increase in stock price. Larry Pinkston, Unit president and chief operating officer, says, "We are pleased with our placement in this ROCE analysis, which confirms our strategy of asset growth while being conservatively financed, both of these resulting in increased shareholder value." For purposes of this analysis, 10% was chosen as the minimum acceptable ROCE an investor would target. Of the 34 oilfield-service and contract-drilling companies included in the study, only four-BJ Services, FMC Technologies, Carbo Ceramics and Unit Corp.-exceeded this minimum rate of return. In general, drilling contractors performed poorly, returning an average ROCE of 3%, while experiencing an approximate 21% decline in stock performance during the studied period. Steven F. Crower is an assistant vice president of investment banking at Hibernia Southcoast Capital in New Orleans. He can be contacted at scrower@hibernia.com. METHODOLOGY Time frame ROCE was calculated for three fiscal years (2001, 2002 and 2003) using quarterly financial data (12 quarters of data) to smooth the year-over-year volatility of financial results, while minimizing loss of data that mergers and acquisitions or a longer time frame would impose. The selected period captures the majority of an energy industry cycle that appears to have become shorter in recent years, as indicated by the three- and four-year cycles seen recently, versus longer cycles of the prior two decades. ROCE Net income before extraordinary charges is used in the analysis because of the inclusion of all operational expenses, taxes and interest expenses. Both the numerator and denominator of the formula were adjusted for minority interests in affiliates from joint ventures. Stock performance The average closing prices of multiple dates were used to minimize stock-price volatility. One data point was the average of the month-end closing price for the first six months of 2001. The second data point is the average of the week-end closing price for the four weeks in February 2004. This accounted for the time lapse between the end of the calendar year and the dissemination of fourth-quarter 2003 financial results to the market. Software programming Southcoast developed specialized programming to perform the large number of calculations required in this study. Approximately 600 stock-price data points and 1,600 financial data points were used in 1,800 calculations performed quarterly and graphed for more complete analysis. Regression analysis Linear regression analysis was used to determine correlations. A correlation coefficient (R2 ) of 1.0 indicates a perfect linear relationship, and a correlation coefficient of 0.0 indicates no linear relationship. Asset write-downs ROCE was analyzed to determine whether asset write-downs or extraordinary events could allow management to manipulate ROCE. The analysis indicates that any write-down of assets inside the three-year time frame of the study actually impaired the company's financial performance. However, if a write-down occurred before the three-year period, ROCE tended to increase because of the reduction of the denominator without a corresponding reduction in net income during the next three years. Accounting method Successful-efforts and full-cost accounting methodologies used by E&P companies were analyzed to see if the choice of accounting methodology affected the calculation of ROCE. Successful efforts allows companies to expense all the costs to drill and complete an unsuccessful well at the time the well is drilled, while full-cost allows companies to capitalize these costs over time. Both accounting methodologies affect the numerator and denominator of the ROCE formula.