Most producers, while enjoying this extended stretch of historically high prices, are writing large checks each month to counterparties for settled hedges. Often referred to as "opportunity cost," sending these checks is painful-this is revenue that would have otherwise gone straight to the bottom line. The piece of mind that should be achieved through hedging often takes a back seat to the uncertainty surrounding the value of the hedge program. To manage opportunity cost and improve hedge performance, several producers now use a statistical methodology to coordinate their hedge activity with short- and long-term strategic goals. By taking a quantitative approach, hedging decisions are taken out of the proverbial "black box" and objectivity replaces guesswork. Hedge programs are developed to optimize performance in ways that match the unique requirements of each producer. Once implemented, hedge effectiveness is monitored to provide feedback that will alert the producer if changes are necessary to ensure strategic success. The end result is a hedge program that instills confidence in senior management. The program provides "insurance" that budgetary targets will be met, while minimizing the opportunity costs associated with hedging. These producers are "logical hedgers." Their primary assumption is that no one can predict the future. Yet, they consistently outperform their peers. How do they control risk and achieve success in an uncertain world? They are like members of the fabled Harvard graduating class where the 5% with a written game plan accumulated 95% of the wealth. The key to their success is that they assess the range of things that might happen, identify problem thresholds that would hamper performance, and manage proactively for success through a range of price outcomes. Here are the questions the logical hedger asks: What are my revenue estimates? What are my minimum revenue requirements? What hedge volume is required to achieve this target? Which instrument should be used? When should the hedge be placed, i.e. do I need to hedge now? Answers to these questions are developed by assimilating data that are internal and external to a company. Required internal data include proved developed producing (PDP) reserve details and budgetary targets; external data includes historical and swap settlements, current market opportunities and estimates for price potential. With data in hand, the logical hedger can develop an accurate assessment of the firm's ability to achieve budgetary goals. Statistical tools quantify the hedge process. Estimates of future revenues are determined for various levels of confidence by modeling price potential, sensitivity to risk and the expected value of selected hedge strategies. Then future revenues are estimated through a statistically predicted range of prices. Hedge strategies are chosen by management for their ability to ensure that revenues will achieve corporate objectives. Modeling price outcomes The statistical methodology used by logical hedgers is widely accepted. Price distributions for every commodity and stock are modeled using the assumption that markets are lognormally distributed. More than 40 years ago, Myron Scholes and Fisher Black used this assumption to develop the first options-pricing model, which remains the foundation of almost every option-pricing model used today. The methodology is used to develop price distributions that allow logical hedgers to understand their risks and make better hedge decisions. Bell curves or normal distributions are characterized by their mean and standard deviations (SD). If measuring the area below a bell curve, 95% of the area is contained within the range of +/- 2 SD. For hedging purposes, this has important implications. If looking outside of this range, one can conclude that 2.5% lies above the +2 SD and 2.5% below the -2 SD. Logically, about 98% of the area under the curve lies above the -2 SD price. Price history is used to develop an understanding of price potential. Because traders like to compare markets, a market's SD is expressed as a percent of the commodity value and called volatility. Volatility describes how much the price can be expected to change in a one-year time period. For example, if natural gas is trading at $5 per million Btu and has 20% volatility, one can estimate with 95% confidence that one year from now, it will be between $3 and $7, which is the +/- 2 SD range. One can also have 98% confidence that it will be above $3. Since January 2000, gas-price settlements have ranged from a low of $1.83 to a high of $9.98. The average settlement during this period is $4.38 +/- $1.62. Meanwhile, the forward curve from May 31, 2004, averaged around $6. It might seem that market participants were optimistic and expect prices to sustain levels equal to +1 SD for the next few years ($4.38 + $1.62 = $6). However, when observing the risked forward curve (-/+ 2 SD range), this optimism did not appear so out of place because the range encompasses much of the price history shown. The risked curve suggests that gas prices are not expected to settle below $3 at any time for the next two years. By converting price to revenues, one begins to develop an understanding of how price potential can affect a company. Revenue estimates for a 30-million-cubic-foot-per-day producer in 2005 might be as low as $38 million (P-98 uses the -2 SD price). If that were to happen, what would it mean to this gas producer? An examination of the crude oil curve for the same date shows that all of the settlements for the past four years fall into the current distribution. Envision the effect of removing backwardation. It would lift the forward curve to $40 per barrel to match the most recent settlement. Backwardation has a negative impact on producers that is generally best managed by hedging as late as possible and as little as one must. A 5,000-barrel-per-day producer can estimate 2005 revenues using a statistically risked price deck of $35 (the curve or P-50) and $18 (-2 SD or P-98) per barrel. Estimates would range between $67- and $32 million. A firm of this size, if it does not hedge, can expect to gain or lose about $1.8 million in revenue with every $1-per-barrel change in oil prices in 2005. Case studies The following are reviews of recent examples where this approach was used-or not used-to quantify the decision process and improve hedge performance. Example No. 1: In this example, a gas producer successfully hedges to ensure achievement of revenue targets. Prices rallied between September and December 2003, and this enabled the producer to increase monthly P-98 estimates. 2004 revenue estimates increased by $40 million to $175 million. Hedges covered 50% of first-quarter and 25% of second- and third-quarter volumes. The gap between P-98 and P-2 narrowed after hedges were implemented. In this example, the effectiveness of the hedges is demonstrated by the lifting of the P-98 level, without dragging down P-2. The high-case revenue this company might achieve remains the same for the possible scenario that prices continue to strengthen. Between December 2003 and March 2004, strong first-quarter settlements and a forward-curve rally raised P-98 estimates for 2004 to $207 million. At this point, the producer has a large cushion over its minimum target of $175 million and is confident that hedging additional volumes will not be needed. Of course, if the market should have a dramatic reversal or the producer's market view should change, the need for additional hedges would be identified. The specific volumes required and the benefit the company achieved would be measured by the improvement to P-98. Example No. 2: In this example, which was pulled from a producer's annual report, the hedge was little more than window dressing and may become a costly mistake. During January 2004, this large independent hedged 32% of its gas, almost 700 million cubic feet per day, with collars where it received a $3.77 floor and provided a $7.31 cap. This increased the producer's P-98 just 2% from $2.45 billion to total $2.5 billion. With the 2004 curve worth $5.30, these puts had less than a 7% chance of protecting anything. In looking at the calls, there was less than 1% chance of them being exercised. On the surface, it appears that a slight advantage was achieved by the producer. However, when evaluating just the winter months, these calls are more valuable and have a much higher chance of exercise. Example No. 3: During March 2004, a small independent was required to hedge 60% of its PDPs when it closed a new lending facility. The producer used a form of collar that increased P-98 to $14.9 million from $11.2 million, or 33%. If swaps had been used, P-98 would have been raised 60%, but most of the upside would have been lost. The hedge used in this example has an unusual return profile. It allows the producer to participate in rallies like a collar, but also provides immediate protection like a swap. The downside is that the protection is limited, but this strategy does give the producer the best of both worlds through 60% of the expected distribution. The company is protected against the first 1 SD (34% probability) of decline in crude prices and participates in the 0.7 SD (26% probability) of price increase. Example No. 4: In another example, the producer had missed an opportunity to hedge crude at $32. This strategy was successfully implemented-it was able to purchase a $32/$34 costless collar in a $31 environment-by first purchasing the puts. Specifically, this producer used the "optionality" of the asset to hedge at better-than-initial prices. Note, this producer never speculated. Each step of the process increased its P-98, or the present value, of his production. Further, the only potential downside to this scenario is if prices were to have gone lower, the producer would have purchased puts he needed! Example No. 5: In September 2003 a producer was prepared to hedge 15 million barrels of crude oil at $27 a barrel to lock in $400 million. Analysis demonstrated that the company had 98% confidence that revenue would be $366 million. Therefore, it only needed to hedge 1.25 million barrels to raise its estimate to $400 million. The producer elected to hedge 5 million barrels to secure the revenue target with a cushion, and decided to monitor the market to determine if additional hedges would be necessary. The 10 million barrels that were not hedged were worth an additional $36 million at year-end. With the continued strength in crude, this decision will wind up saving this company more than $100 million. These case studies illustrate how statistical analysis can facilitate the coordination of hedge activities with corporate goals, making it possible to achieve consistent success in an uncertain price environment. M Wayne Penello is president of Risked Revenue Energy Associates, a Houston-based consulting firm. He is founder of the Risked Revenue Hedge Program and can be reached at 713-807-1920.

Comments