Risk Management

Risk management

Deregulation and competition increase the volatility of energy prices. The more volatile an energy market is, the riskier it is for firms doing business in that market. Energy traders call this risk market risk, and some now quantify it using measures other than, but based on, the original one—Value at Risk Value at Risk: Variations on a theme

alue at Risk (VaR) is Since then, some trading during a given period of time a number that the firms have begun using what under normal market condi- chief executive of they consider better mea- tions. The final VaR commu- any energy trading sures of market risk, all of nicated to top management V company should which are based on VaR. at the end of each day is know before leaving the office Three of these new measures aggregated from the individ- each day. It indicates how are Profit at Risk (PaR), Earn- ual VaRs of the company’s much money the company ings at Risk (EaR), and Cash different trading desks and could lose if market prices Flow at Risk (C-far). The mea- portfolios. In the first section, move wildly before the com- sures are described in the Dr. Carlos Blanco describes pany can close its position. four sections that follow. the three main techniques for Originally used in finance, Specifically, VaR measures calculating VaR: analytic VaR, VaR was adapted years ago the worst expected loss (for Monte Carlo, and historical by energy trading compa- a portfolio) to a specified con- simulation. nies to meet their needs. fidence level (usually 95%) Just how useful is VaR for

12 Global Energy Business, May/June 2001 Risk management energy companies? In the second section, Chal Barn- Calculating and well points out its weak- nesses. He says that if energy companies use VaR the using Value at Risk way that banks use it, they could grossly underestimate ost risk measurement level for the set. their risk exposure. The bet- methodologies are based There are three main methodologies ter measure of market risk on the analysis of a set of for calculating Value at Risk (VaR): vari- scenarios that describe pos- ance-covariance (also known as analytic that his company has come M sible future “states” of the VaR), Monte Carlo simulation, up with is a variation of VaR world. Each of the method- BY DR. CARLOS and historical simulation. The BLANCO called PaR. ologies makes different three are complementary, but Also recognizing the limi- assumptions about the possible evolution each offers a different view of portfolio of markets, but they follow a similar risk. Ideally, a firm would use all three tations of VaR are Dr. Gary approach towards measuring market methods to obtain the most accurate pic- Dorris and Andy Dunn. In risk: First calculate a profit or rev- ture of the market risk it faces. the third section, they intro- enue for each scenario, and then aggre- duce another measure of gate those results to form a distribu- Variance-covariance tion and extract the mean and variability market risk—EaR—that they of the profit, portfolio value, or revenue The most commonly used of the VaR say is more suitable for physical-asset-intensive energy 1. Monte Carlo simulation companies. The article includes an example of how Current forward Volatility and managers of a hypothetical curves, other prices rrelation information gas-fired merchant power plant might use EaR to limit their company’s market risk Scenario generation engine 4 exposure. 1 2 Finally, in the fourth section, Louis Guth and Kristina Sepetys explain how their New prices for scenario 1 for scenario 2 scenario 10,000 company’s patent-pending C-far model can provide an Hypothetical MTM for each price scenario answer to a question that energy trading firms often ...... ask: “What is the probabil- –$645,334 –$243,000 $543,000 ity that this year’s cash flow 160 will be inadequate to fund 140 Monte Carlo simulation VaR our strategic investments?” 120 The answer takes the form 100 of a risk profile, which the 80 model generates by taking 60 into account the different types of risk to which a com- occurrences of No. 40 pany is exposed. 20 0 -1.258 -1.146 -1.034 -0.922 -0.810 -0.698 -0.586 -0.473 -0.381 -0.249 -0.137 -0.025 0.087 0.199 0.311 0.424 0.536 0.648 0.760 0.827 0.984 —Anne Ku Portfolio profit or loss, $1 million

Global Energy Business, May/June 2001 13 Risk management 2. Historical simulation ical profits and losses, or P&L, of the firm’s portfolio under each scenario are converted into a histogram of Current Data base with historical forward expected profits and losses, from which market prices curves, exchange rates, interest rates VaR can be calculated (Figure 1). One advantage of Monte Carlo simulation is that it does not assume that 2 portfolio returns are distributed normal- ly. Another is that it is a forward-look- New set of prices New set of prices New set of prices for scenario 1 for scenario 2 ...... for scenario 100 ing assessment of risk that takes into 1 account options and non-linear posi- 3 tions. However, the methodology requires the use of a correlation and ...... volatility matrix to generate the random scenarios, and that makes it computa- tionally intensive. It also requires the company to have pricing models for all the instruments in its portfolio. Estimated P&L Estimated P&L Estimated P&L 4 –$1,215,334 –$443,000 $643,000 Historical simulation 160 Historical simulation refers to the 140 Historical simulation VaR The estimated P&L for each scenario is the process of calculating the hypotheti- 120 difference between the cal distribution of profit and losses MTM of the portfolio of a portfolio based on how it would 100 under each scenario and the current MTM. If have behaved under several hundred 80 you group all possible scenarios in the past. Its advantages are 60 P&Ls, you can get that it does not use estimated vari- the P&L distribution ances and covariances, and does not No. of occurrences of No. 40 assume anything about the distribution 20 of portfolio returns. However, the big 0 -1.258 -1.146 -1.034 -0.922 -0.810 -0.698 -0.586 -0.473 -0.381 -0.249 -0.137 -0.025 0.087 0.199 0.311 0.424 0.536 0.648 0.760 0.827 0.984 disadvantage of historical simulation Portfolio profit or loss, $1 million is its assumption that future risks are much like past risks, and that is less calculation methodologies is based to derive a “synthetic” portfolio of assets frequently the case in today’s fast- on an analysis of the volatilities of, and held. The synthetic portfolio is made up changing energy environment. correlation among, the different risk of (cash flow) positions in the risk fac- To calculate VaR through historical exposures of the firm’s portfolio. tors, or “vertices,” whose volatilities and simulation, one needs two things: a data The calculation of analytic VaR is a correlations are known. base with historical prices for all the two-step process: The purpose of cash flow mapping is risk factors to be included in the simu- Select a set of market risk factors to find the “best” replication of a finan- lation, and pricing models to re-evalu- and systematically measure actual price cial exposure for the purpose of measur- ate the portfolio for each price scenario levels, volatilities, and correlations. ing its risk in conjunction with the firm’s (Figure 2). Historical simulation could Put the firm’s exposures into a other exposures. The most difficult part be considered a special case of Monte form that can be analyzed using risk of this step is defining a set of risk fac- Carlo simulation in which all scenarios factor information. This is called cash tors that is small enough to be manage- are defined before the event according flow mapping. able, but comprehensive enough to cap- to the past behavior of market prices. In Market risk factors refer to anything ture all the firm’s risk exposures. Once the case of electricity and over-the- that affects the value of the portfolio. the cash flow map has been created, one counter energy markets, it is quite diffi- Prices are the most common market fac- need only perform basic matrix manipu- cult to calculate VaR using historical tors, but you could also include in the lation to calculate the VaR of a portfolio. simulation because price histories are analysis non-market-related risks—such hard to come by. as volume and weather. To be compati- Monte Carlo simulation ble with the available risk factor data, A new way to calculate VaR every instrument in a portfolio needs to Monte Carlo simulation generates ran- be reduced to a collection of cash flows dom pricing scenarios. The hypothet- Risk managers are primarily concerned

14 Global Energy Business, May/June 2001 Risk management with the risk of events that are very 3. Extreme value theory applied to value at risk unlikely to occur but could lead to 1.6 catastrophic losses. Yet traditional EV-VaR vs. normal VaR 1.4 VaR calculation methods tend to ignore Historical extreme events and focus on risk mea- 1.2 simulation sures that accommodate the entire results empirical distribution of returns. This 1.0 is a problem, because it is extreme 0.8 events—like a large market move—that VaR based on extreme value distribution VaR based produce the largest losses. 0.6 on normal Using stress tests and scenario analy- distribution ses to simulate the changes in the value 0.4 Cumulative probabilities, % of a portfolio under hypothetically 0.2 extreme market conditions is no solu- 0.0 tion, because they cannot explore all –35 –30 –25 –20 –15 –10 –5 0 possible scenarios. What’s more, such Loss, % analyses do not indicate how likely it is that extreme events will occur. companies has been limited to a pas- is more complicated—and interest- The problems resulting from extreme sive role—for reporting rather than ing—than risk management in most events are not unique to risk manage- prescriptive purposes. Now, howev- other fields because electricity and gas ment; they also arise in disciplines such er, firms in dynamic industries— production and trading are physical as hydrology and structural engineer- such as energy—are beginning to use processes. ing, where extreme events can have it for more proactive purposes—for As energy markets become more devastating consequences. Researchers risk management rather than just risk complex, energy firms are embracing and practitioners in these fields handle measurement. Active VaR can iden- risk management systems and proce- the problem by using extreme value tify business activities that incur too dures, such as Value at Risk, to become theory (EVT). EVT is a specialist much risk relative to their level of more competitive. branch of statistics that derives general return. It can also point out which properties of the tail, or extreme end, of assets, business processes, and activ- a distribution by making the best possi- ities are increasing or decreasing the Dr. Carlos Blanco is manager of global support and educational services at ble use of a limited set of its realized corporation’s overall risk exposure. Financial Engineering Associates, extreme values. By focusing on the Most risk professionals agree that (www.fea.com) Berkeley, Calif. extreme tail of a distribution, VaR can risk management in the energy industry be estimated with a confidence of greater than 95%. The difference EVT makes to VaR estimates is illustrated in Figure 3, which shows the tail of the West Texas Profit at Risk: Intermediate (WTI) daily return distribution from 1983 to 1999. The dots indicate the actual extreme return observa- More realistic than tions, the continuous line by the right vertical axis represents the tail (assuming that logarithmic returns follow a Value at Risk normal distribution), and the other continuous line represents the tail of an extreme value distribution fitted to ompanies that generate or Volume risk these data. The message this figure con- deliver electricity, or sell veys is that the confidence level of load-following ancillary ser- The financial risks estimated by typ- EVT-calculated VaRs is much higher Cvices or retail load services, ical VaR calculations are wholly relat- than that of VaRs calculated using tradi- need a more accurate and ed to market price movements. tional methodologies. robust risk-measurement met- BY CHAL However, the real and increas- ric than the “pure” form of BARNWELL ingly complex energy industry Toward risk management Value at Risk (VaR). Profit at Risk engenders additional financial risks (PaRTM) is a more useful measure for a that VaR cannot measure. The most Traditionally, the use of VaR within variety of reasons. important of these is volume risk,

Global Energy Business, May/June 2001 15 Risk management which occurs whenever a power plant 1.Breakpoint has an unplanned outage, customers use more power than expected, or a transmission line fails. 150 Volume risks—and their impact on Relatively 80 Delivery price value and profit—are not caused by stable could vary forward market price movements, but rather by 60 significantly prices from the physical problems, such as those men- forward price tioned above. If anything, the converse expectation is true: Market price movements may 30 result from volume risks. In fact, the Feb 00 May 00 Sep 00 Feb 01 June 01 biggest and most shocking market price Break point movements in power markets can all be traced back to physical volume risks. change dramatically. After a certain the correlations between risk factors and date, or breakpoint, prices become dri- the volatility of these factors. The rela- Liquidity risk ven by less benign factors such as tionships are derived from econometric weather, the short-term imbalance estimates based on historical data and/or Companies that use VaR to measure the between supply and demand, plant out- inferred from current market variables— impact of market price movements on ages, and transmission system bottle- options prices, for example. the value of their asset portfolios typ- necks. The environment becomes much The PaR approach considers both the ically rely on a “stop loss” strategy. more volatile. For energy trading firms, volatility and shape of the forward price They believe that they can quickly VaR remains a valid way to measure curves observed historically during the close out their position to limit the their risk exposure before the break- delivery period, as well as projected damage from negative market price point date. However, for many other price curves and volatility for that time movements. However, the time required energy companies—including genera- frame. The PaR metric, then, reflects to close a position depends on market tors, retail distributors, and entities the entire exposed profit of a portfolio liquidity—but liquidity changes over involved in complex marketing transac- position, not just the change expected in time and fluctuates in response to sud- tions with load following optionality— it for a brief period of time. den shifts in market prices. In imma- PaR offers a way to more accurately ture power markets, price spikes can measure their risk. PaR in practice be extreme and cause severe market liquidity problems. PaR in theory Short of selling a portion of a gener- In practice, therefore, closing a posi- ating unit, a company has no way to tion is neither as easy nor straightfor- The biggest difference between PaR effectively close out the position rep- ward as in theory. Because many real- and VaR is that the former assumes resented by a plant without assuming world contracts do not comply with mar- that positions will be taken through to potentially more risk than is hedged. ket standards, they are not easy to sell or delivery rather than closed out prior The plant will be subject to forced out- trade. The alternatives—selling off gen- to the breakpoint. Because this mea- ages, transmission constraints, generation assets or portfolios of retail cus- sure takes into account the full finan- eration derates, and many other volu- tomers—are equally unrealistic. cial risk of highly volatile spot prices, metric variables that are uncontrollable. it is a far more realistic approach. In many cases for strategically locat- Physical delivery and The foundation of PaR is simulation- ed facilities, forced outages can drive breakpoints based modeling, which basically tests spot prices to abnormally high points the entire range of risks affecting spot at exactly the wrong time for the What makes assessing risk in power prices at the time of delivery. The model hedger—when the facility is down. In markets complex is that market prices follows a pseudo-Monte Carlo approach these circumstances, a VaR measure change drastically as the date of phys- that re-bases spot price simulations to based on a short holding period of two ical delivery approaches. Prior to sev- match forward curve prices and expect- or three days would give the impres- eral months before that date, price ed spot volatility. The Monte Carlo sim- sion that a plant’s risk is minor when behavior is driven by relatively benign ulation method calculates the change in it is actually quite substantial. factors—such as economics, forward the value of the portfolio using a sample For example, calculating the VaR of a fuel prices, and the market’s anticipated of randomly generated price scenarios Cinergy 500-MW gas-fired plant on generation level. This is considered the that are assumed to be equally probable. May 30, 2000, would produce a figure holding period. The approach requires making assump- of $5.2 million for the June through However, as the date of delivery tions about market structures, the sto- September period. But using the PaR approaches, the drivers of market prices chastic processes the prices follow, and metric, the calculation of total profit

16 Global Energy Business, May/June 2001 Risk management exposure for this same unit (based upon situation was the volumetric uncertainty derivative contracts, and to that of a com- a 95% confidence level) would produce of load, which effectively prevented bined portfolio of derivative contracts a very different figure—$69 million. utilities from closing out their retail and the merchant plant. What this means is that leaving the plant positions. And to make matters even For the price simulation, assume that completely unhedged for the premium worse, regulatory rules made it illegal derivative prices follow geometric summer months and selling its output on for them to hedge their positions sub- Brownian motion with a monthly term a daily basis would have resulted in an stantially. structure of volatility and drift, and spot erosion of mark-to-market (MTM) prof- The PaR metric would not have prices follow a Markov regime-switch- itability of $69 million by the end of helped California’s utilities out of their ing model with each regime character- August. regulatory dilemma. But it might have ized as having geometric Brownian The shortcomings of VaR—in particu- flagged the potential for eroding profits motion with mean reversion. The lar its assumption that past events are in a simulated environment and provid- regime-switching model for spot prices reliable indicators of the future—are ed an accurate measure of their risk allows for the large discontinuous perhaps most glaring if one considers exposure. jumps observed in electricity prices. the California crisis. No amount of his- Figure 1 depicts a single simulation torical data on wholesale prices could of forward and spot prices for power have predicted what happened in the Chal Barnwell is vice president of the during a typical summer in North spring of 2000—events that haven’t Houston operations of KWI (www.kwi.com), America. Spot prices follow the highly London. recurred since. Further complicating the volatile purple line at the front of the graph. Forward contracts are less

1. Sample simulation of Earnings at Risk: power prices Better for asset owners

ccounting rules prohibit ener- the profit function. The distribution of gy companies with portfo- profit captures the upside potential—as lios of physical assets well as the downside risk—of BY DR. GARY from marking these variability of market prices, as A DORRIS AND assets to market. Nor can these well as the operational charac- ANDY DUNN Table 1. Generation companies liquidate or acquire teristics and reliability of physi- dispatch inputs and their assets as quickly as companies cal assets. A well-constructed EaR outputs with purely financial portfolios. For model assesses the impact of alternative Operational inputs Total these two reasons, energy companies derivative trading strategies on both Load...... 130 are embracing Earnings at Risk (EaR) tails of the profit distribution curve. Heat rate ...... 13,653 as a more appropriate measure of mar- Minimum uptime, hours ...... 2 Minimum downtime, hours ...... — ket risk than Value at Risk. EaR in action Startup costs, $/MW of capacity ...... 1 Startup time, hours ...... 1 How might EaR be used in the ener- Shutdown cost, $/MW of capacity . . . 229 EaR defined Shutdown time, hours ...... 1 gy industry? Consider the hypotheti- Ramp-up rate, MW/hr ...... 3 Calculations of EaR follow calcula- cal case of E-Corp., a U.S. firm that Rampdown rate, MW/hr ...... 15 tions of profit, using the following owns a 130-MW gas-fired merchant Maximum starts per month ...... 40 Maximum continuous run time/ equation: plant and has a marketing/trading affil- month, hours ...... 744 Profit = Spot price revenues generat- iate that does both spot and forward Expected forced outage rate...... 0.10 ed by the organization’s assets transactions in electricity and natural Expected outage duration ...... 2 Variance in outage duration ...... 1 Ð cost of production, gas. E-Corp.’s objective is to manage Outage minimum, days ...... 1 + hedge and other instrument its exposure of earnings to market risk Outage maximum, days...... 120 payoffs prior to delivery, over a 12-month period. First, simu- Summary output Total + hedge and other instrument late prices, and then estimate the EaR Generation, MWh ...... 773,500 Gross revenue, $1,000 ...... 120,157 payoffs during delivery. of E-Corp.’s unhedged portfolio. Next, Net revenue, $1,000 ...... 54,043 A Monte Carlo simulation can pro- compare the EaR of this unhedged Net revenue, $/kW ...... 415 vide an estimate of the distribution of portfolio both to that of a portfolio of Capacity factor, %...... 68

Global Energy Business, May/June 2001 17 Risk management volatile and turn flat during delivery. Of Statistics: 2. Distribution of net revenues course, this is only a single simulation EaR (C.1. 95%) 11,109,200 100 of prices; a fair assessment of risk Mean 54,034,300 90 requires several thousand simulations. Maximum 74,851,500 Minimum 32,599,900 80 Given a series of price simulations S t d . d e v. 7, 4 97,6 9 0 and the operating attributes for the 70 plant, the dispatch of electricity as well Bin data: Level Net revenue 60 as the corresponding revenue informa- 0 32,599,900 50 tion can be simulated for the year by 1 37,721,800 mapping each hour of each day’s spot 5 42,925,100 40 price for power and natural gas to the 10 44,797,200 30 potential generation of electricity. Table 20 47,002,200 occurrence of Probability 30 49,445,100 20 1 tabulates the operational attributes of 40 51,543,600 10 this plant, its expected generation, 50 53,826,900 60 56,053,800 0 capacity factor, and expected revenues. 3.26E7 3.68E7 4.11E7 4.95E7 5.8E7 6.43E7 6.85E7 7.27E7 70 57,773,800 However, Table 1 provides only a Net revenues, $ 6.01E7 portion of the information needed for effective risk management and budget- of owned assets, commercial contracts, set the volatility of the earnings of the ing. The rest of the data needed must and/or hedge positions. Suppose that in power plants. When these portfolios are come from an examination of the distri- addition to the merchant plant, E-Corp. aggregated into one exposure, overall bution of net revenues. Figure 2 depicts has a strip of natural gas forward pur- risk is reduced to $3.5 million, based on the probability density function (PDF) chases (1 million mmBtu) and electricity the offsetting profits of the plant and the and other key statistics for the sales (130 MW). A typical risk manage- derivative contracts. unhedged power plant. The statistical ment activity would be to measure the The example clearly shows the hazards summary on the left indicates that the VaR or EaR of this portfolio alone with- of not fully considering all assets that are plant’s expected net revenues are esti- out considering the generation asset. subject to market risk in risk analysis. If mated at $54 million, and that its However, the generation plant has the E-Corp. had measured the portfolio of Earnings at Risk figure is $11.1 million. potential to offset some of the risk of this forward contracts in isolation, it would In other words, you would expect that portfolio of forward purchases and sales. have inaccurately estimated market risk, the plant will generate $54 million in Table 2 illustrates that both the gener- and that could have led management to net revenues, and it is 95% certain it ation plant in isolation and the portfolio pursue an erroneous hedge strategy. The will generate at least $42.9 million. of forward contracts carry significant example underscores the need for energy Once the numbers have been deter- levels of EaR ($11.1 million and $13.9 companies to apply sophisticated model- mined, you can measure the market risk million, respectively). In the profit equa- ing techniques to all of its market-price- with respect to the earnings of a portfolio tion for EaR, the derivative contracts off- sensitive assets using a method that estimates risk to earnings. Table 2. Risk-reducing effects of aggregating From VaR to EaR generation with forwards Since they came out of the world of Generation plant alone Forward contracts alone Combined portfolio EaR (C.I. 95%) . . 11,109,200 EaR (C.I. 95%) . . 13,853,430 EaR (C.I. 95%). . . 3,518,400 finance, methods of measuring market Mean ...... 54,034,300 Mean ...... 2,091,030 Mean ...... 56,209,200 risk have evolved in sophistication Maximum . . . . 74,851,500 Maximum . . . . 24,442,800 Maximum . . . . 63,966,700 and robustness, from basic VaR to Minimum . . . . . 32,599,900 Minimum . . . . (18,228,000) Minimum . . . . . 50,635,900 measures, such as EaR, that integrate Std. dev...... 2,351,140 Std. dev...... 7,497,690 Std. dev...... 7,861,020 the risks of financial instruments and Bin data Bin data Bin data those of physical assets. The benefits Level ...... Net revenue Level ...... Net revenue Level ...... Net revenue of using such advanced methodolo- 0 ...... 32,599,900 0 ...... (18,228,000) 0 ...... 50,635,900 gies extend from improving a compa- 1 ...... 37,721,800 1 ...... (16,786,800) 1 ...... 51,047,100 ny’s leverage capacity to stabilizing 5 ...... 42,925,100 5 ...... (11,762,400) 5 ...... 52,690,800 volatility in its earnings to enhancing 10 ...... 44,797,200 10 ...... (7,955,730) 10 ...... 53,384,000 20 ...... 47,002,200 20 ...... (4,674,190) 20 ...... 54,248,900 its stock-price-to-earnings ratio. 30 ...... 49,445,100 30 ...... 49,445,100 30 ...... 54,954,900 40 ...... 51,543,600 40...... 193,180 40 ...... 55,537,500 Dr. Gary Dorris is CEO, and Andy Dunn is 50 ...... 53,826,900 50 ...... 1,577,460 50 ...... 56,050,700 vice president of risk management of 60 ...... 56,053,800 60 ...... 4,574,250 60 ...... 56,561,400 e-Acumen (www.e-acumen.com), San 70 ...... 57,773,800 70 ...... 6,001,630 70 ...... 57,106,300 Francisco.

18 Global Energy Business, May/June 2001 Risk management

lish whether a company has adequate cash reserves to service its debt in the Cash Flow at Risk event of an earnings jolt. What’s more, because it can inform decision-making about purchasing insurance and other for non-financial hedging strategies, C-far could become a benchmark for corporate risk man- companies agement. Tailored for the industry

ash Flow at Risk (C-far) fy reported risk in securities liti- For power companies, the C-far pack- By Louis Guth is a patent-pending diag- and Kristina gation and disclosure docu- age comprises the model and a data nostic developed by Sepetys ments. Investment bankers and base of 10 years’ worth of quarterly CNational Economic venture capitalists will find it a cash flow and other financial data Research Associates (NERA) econo- useful method for analyzing investment for approximately 100 electric utili- mists. This analytic model is designed opportunities. ties. The model includes a statistical to forecast earnings volatility one methodology for transforming these quarter to one year ahead. It may be C-far vs. VaR data into peer-benchmarked risk mea- used to enhance planning and capital surements. It can construct tailored, investment, as well as insurance and Banks, insurers, investment firms, and forward-looking cash flow distribu- hedging strategies. It derives from a other financial companies have long tions for a utility’s quarter-ahead and broad statistical analysis of earnings used Value at Risk, or VaR, to manage year-ahead horizons based upon large results and other factors for the entire portfolio risk. VaR measures how much samples of cash flow shocks experi- universe of publicly held non-finan- the value of financial assets—foreign enced by a set of peer-group firms. cial companies. C-far starts with the currency, equities, commodities, These samples are large enough to logical premise that risk reveals itself bonds—will drop in a day or a week allow making relatively precise state- in non-financial companies as devia- tions from cash flow. C-far promises to enhance financial What and who it’s for strategy and long-term investment planning The introduction of the C-far model has coincided with what has become a if they are affected by a market rever- ments about the probability of one very big season for unanticipated earn- sal. The C-far model can be thought in 10-year or one in 20-year bad out- ings shortfalls at some of the country’s of as a form of VaR for measuring comes. The distributions also allow largest companies, including many aggregate risk against a company’s quantification of the probability of utilities. The model can provide answers cash flow. Whereas the VaR method specified adverse cash flow shocks— to two questions that should be fore- takes a bottom-up approach to quan- for example, “What is the likelihood most in the mind of any CFO or high- tify the risks caused by individual that cash flow falls by 50% in the level manager concerned with accu- financial assets, C-far looks directly next year?” rately assessing value: at cash flow under the assumption that Such information is of great interest “What is the probability that this all risks to a corporation are manifest to investors, who are naturally con- year’s cash flow will be inadequate to in shocks to expected earnings. cerned about volatility in reported quar- fund our strategic investments?” Because the C-far model creates for- terly earnings. By disclosing the results “What is the worst thing that can ward-looking probability distributions of a C-far analysis to investors or secu- happen to the company financially in for a company’s cash flow, it provides rity analysts ahead of time, a company the next quarter or year?” a measure of the likelihood of rare neg- can help put earnings shocks into a The C-far model considers every con- ative events that could produce a sig- credible, objective, peer-benchmarked ceivable type of risk to which a compa- nificant drop in earnings. As a tool for perspective. ny can be exposed and produces a risk corporate treasurers and CFOs, C-far profile that can help answer these criti- promises to enhance financial strategy Louis Guth is senior vice president, and cal questions. Its applicability goes and long-term investment planning. Kristina Sepetys is senior consultant (San beyond volatile industries—such as But the model can also be used to help Francisco office) at National Economic energy, manufacturing, and high-tech companies assess their capital structure Research Associates (www.nera.com), White technology. Lawyers can use it to clari- and credit-worthiness. C-far can estab- Plains, N.Y.

Global Energy Business, May/June 2001 19 Global Energy Business is written to meet the needs of the time-constrained energy executive. We examine energy issues as part of the global economy. We explore the impact of international and local politics, economics, and politics on the energy business--then deliver concise market intelligence to our readers. We profile pacesetting executives and innovative companies. Part of our "big picture" perspective is recognizing that our readers don't operate in a vacuum, that's why each issue includes provocative market and financial commentary, travel, and leisure articles; and a column that looks at the lighter side of business.

Please complete the form below to receive a FREE subscription to Global Energy Business.

1. [ ] Yes, start/renew my FREE subscription to Global Energy Business. [ ] No, I do not wish to receive Global Energy Business.

2. Name:

E-mail address:

Title:

Company:

Address:

City:

State or Province:

ZIP or postal code:

Country:

Phone:

Fax:

3. Primary Job Function (check only one): [ ] General or Corporate Management - A [ ] Library or Company - B [ ] Other (please specify): 4. My company is a (check only one): [ ] Primary energy supplier - A [ ] Utility - B [ ] Developer/independent power producer - C [ ] Energy services provider - D [ ] Consultant, contractor, constructor - E [ ] Energy marketer, trader, broker - F [ ] Commission, association, or government - G [ ] Law, financial services, or insurance firm - H [ ] Other (please specify):

4a. In question 4 above, if you checked PRIMARY ENERGY, what does your company provide (check ALL that apply) [ ] Coal - A [ ] Oil - B [ ] Gas - C [ ] Nuclear Fuel - D [ ] Other (please specify):

4b. In question 4 above, if you checked UTILITY, what services does your company provide (check ALL that apply) [ ] Electric - A [ ] Gas - B [ ] Water - C [ ] Telephone/communications - D [ ] Other (please specify):

State or Province of Birth:

(This personal information is required by auditors to receive a free subscription and will be kept strictly confidential.) For more information about our privacy policy visit http://www.globalenergybusiness.com/privacy.html

If you received this form electronically, please email to [email protected]

If you did not receive this form electronically please sign and date below and

Fax to 1-609-426-7797

Signature Date .