In the survival risk jungle

In the survival risk jungle

Keers, Greg

When there’s danger from volatile spot markets at one side of your business and vulnerability in forward markets at the other, you can tame both by using two risk methodologies–value-at-risk and profit-at-risk–for enterprise-wide risk management.

You can categorize competitive energy market players by their position on a sliding scale-on one end, there are companies that focus most of their energies on trading, and on the other end are those that focus on their physical assets and retail operations.

Those companies’ risks depend on their business strategy. How they manage those risks– and measure and account for them-also depends on what kind of business they are in.

In the risk management arena, there has been a convergence of two historically distinct types of risk management-financial and corporate. The former is familiar in trading operations with accepted standards for business organization, processes, and statistical quantitative techniques. The latter draws upon practical business know-how, sharp negotiating skills, and financial planning and forecasting techniques.

Competitive pressures have forced these two domains to converge in order to provide a consistent, integrated approach, no matter what a company’s focus is. And the pressures are growing due to investors’ and analysts’ demands for transparent and stable financial performance.

At the heart of this issue is everyone’s genuine concern about accounting and risk management techniques and a need to understand the source of a company’s earnings and associated uncertainties. As a utility assesses and manages the risks attached to its particular business model, it needs to find the right path between the risks and the way it accounts for them.

Where Are You on the Scale?

A highly trading-centric market player operates in energy markets much like a bank does in capital markets-it aims to make profits from speculative positions. Within the constraint of risk capital limits, traders execute their strategies in order to maximize mark-to-market gains-that is, adjusting accounts to reflect daily price changes. If these organizations have physical energy operations such as production, transportation, or retail sales, then wherever possible these are connected to equivalent trading transactions-they are simply considered as part of the trading portfolio. The primary profit driver for this type of market player remains the performance of the trading operation and its ability to deliver strong and stable results.

On the other end of the scale, the market player focused on retail operations and physical assets seeks to operate them as effectively as possible to deliver a return on investment over many years. These players invest in more projects depending on operating margins secured through long-term forward contracts for energy commodities. While the most sophisticated players also have highly active trading operations, the primary objectives of the retail operation are to

* help extract as much value from the physical operations as possible and

* support the management of the risks most important to a retailer and operator, namely, market price and volume risks.

The primary profit driver for this type of market player is the competitiveness of the core physical operations. Flexible, low-cost assets and retailing operations are the key to strong and stable earnings for the asset/retail-centric company.

You can safely say that all competitive market players position themselves somewhere between the two extremes. However, in the wake of the Enron debacle business models are certainly inclining in the physical-asset and retailcentric direction.

Earnings and Accounting

Through the use of questionable accounting treatments, Enron raised accounting policies to the status of a front-page issue, debated and discussed by businesspeople everywhere. At the center of the debates are two alternative accounting policies for reporting earnings: markto-market (MTM) and accrual.

Regulated energy companies’ reported earnings almost exclusively have been based on accrual accounting, which is (relatively) straightforward. A company that generates power, for example, would derive the costs, revenues, and subsequent earnings over a financial reporting period by analyzing the costs and revenues directly attributable to physical operation over that period. Likewise, a retailing operation would derive revenues from energy sales and the costs of associated energy purchases or production by analyzing the costs and revenues directly attributable to the financial reporting period.

MTM accounting can be used to book profit or losses from forward energy periods to the current accounting period. Clearly, MTM accounting is aggressive when used for MTM profits and conservative when used to bring forward MTM losses.

All organizations derive their earnings from different mixes of accrual- and MTM-based earnings. But the mixture depends on where the company is on the scale. Traders favor MTM because of their reliance on forward markets; those with physical assets may tend toward accrual accounting. A recent analysis by ABN AMRO suggests that shareholder-owned power and gas companies currently attribute between 5 percent and 30 percent of their earnings to MTM accounting. The Enron episode has exposed the shortcomings of using MTM for long-term, highly complex deals in markets that are not particularly liquid. But MTM is useful, and even necessary, for speculative trading of liquid standard products over a short time horizon.

Exposures, Risks, and Pitfalls

MTM earnings calculations use forward market prices (forward curve) and volatilities (volatility curve) as the key inputs-basically, the market price at the time of delivery, and the risks that could affect the price. Though objective market data may be difficult to find, especially for long– dated positions, the concept is to value your current (that is, open forward) positions against forward market price.

Since the volatility curve represents the degree that the forward market prices are likely to change, it is clear why the volatility of forward market prices is the key risk driver for MTM earnings. If a company has significant open positions in highly volatile markets, MTM earnings will be unstable. There is a compounding effect, too. For example, an adverse change in forward market prices may reduce the MTM profit of positions already accounted for in a previous reporting period (though you deal with this by booking a loss in the current reporting period). Still, in essence, all MTM earnings for unrealized positions continue to be exposed.

Accrual earnings are calculated with reference to the market prices and positions as energy is delivered through the current reporting period only. The spot market price and the position in that market are the key inputs to the calculation-so, spot volatility rather than forward market volatility is a key risk driver for accrual earnings.

Another major driver is volume risk, which is the uncertainty in spot market positions due to unpredictable events such as outages or changes in energy demand. Spot market prices in power and gas markets have proven extremely volatile and subject to non-normal behavior such as price spikes. Combine this with the uncertainty of spot market positions that results from volume risk, and you get a severe risk management challenge-the price is highly uncertain and exposure to it is subject to random change.

The volume risk challenge is made even more difficult when there are adverse correlations between spot price movements and market position changes caused by the volume risk.

For example, energy retailers are more likely to experience spot price spikes at times when their customers consume more energy than expected-and the financial exposure increases. Alternatively asset operators are more likely to experience spot price spikes at times when they have outages, also causing increased financial exposure to the spot price spike.

Optimism and Pessimism

Value at risk (VAR) is a statistical measure of risk exposure-the potential loss from an adverse event in the market. There are many different methods to calculate VAR, but the core concept is that it is a measure of how much the MTM value of open positions might decrease before the market positions could be closed. unit therefore provides a snapshot of how much of accumulated MTM earnings-those that are booked but continue to be exposed-could be wiped out before final delivery.

VAR is the most suitable methodology to capture the risk of the shorter holding– period and to set limits for traders.

Two key parameters are in the calculation.

* Confidence interval-the degree of confidence (typically 90-99 percent) that the user chooses to use for the risk measure. The higher the confidence interval, the higher the resulting VAR.

* Holding period-the number of days the user assumes it would take to close market positions-e.g., three or ten trading days. This is set to reflect the market liquidity: More liquid markets allow positions to be closed more quickly and are therefore linked to shorter holding periods.

Most vAR methods use assumptions that are quite optimistic when applied to energy markets, and that results in a high confidence interval. Here are some common assumptions:

* All elements of the portfolio can be accurately mapped to liquidly traded instruments.

* Market volatility is stable over time.

* Liquidity is reliable and stable over time (i.e., we can definitely close all our market positions within the holding period).

* Forward market price and volatility information is available.

Forward market volatility is a major input to unit calculation. This may either be input in the form of a volatility curve or be directly inferred from historical or simulated market prices.

Where VAR directly measures the risks of achieved MTM earnings being lost over the next few trading days, PAR (profit at risk) focuses on risk to accrual-based earnings over specific future operating periods. PAR is a measure of how much of a business’s accrual– based earnings for a future period might fall below expectations. Typically PAR analysis is aligned with key financial reporting periods, such as the current or upcoming quarter or financial year.

Similar to a vAR calculation, one key PAR parameter is the confidence interval-in this case, the degree of confidence that the user chooses for the risk measure is typically 80-95 percent. The second parameter is the future reporting period, over which a company analyzes the risks to earnings (e.g., the calendar year 2004).

The PAR confidence interval is lower than VAR because the assumptions for PAR are more conservative. One such assumption, for example, is that market liquidity is unreliable and positions will need to be taken through to delivery.

Many would say that is even pessimistic. Yet in the wake of recent market shocks, the liquidity of power and gas markets has been drying up. In fact, market shocks are just the sort of event that risk management defends against– and it is probably useful to the risk manager not to rule them out.

The Best Defense Is Preparedass

Unsurprisingly, the risk management approach best suited to trading-centric business units is largely the same as that of a capital markets trading operation. (See Figure 1.) The basic concept is to allocate a certain amount of risk capital to underwrite the trading operation and then create and track a set of risk limits to ensure that the capital is not exceeded. The traders set about maximizing their trading profits until a limit is approached or breached. In the event of a breach, the typical reaction is to reduce financial exposure by reducing some market positions.

VAR and credit risk measures are the cornerstones of the control of risks in the trading operation. VAR is helpful this way in that it generally looks a few days into the future. In highly trading-centric organizations, where every trader focuses on his own performance, it is essential to manage limits on such a short-term basis for individual traders.

When using PAR to manage the risks to accrual-based earnings, it is typical to look much further into the future. With the volatility of power and gas markets, especially the spot and ancillary markets, a short-sighted asset/retailcentric player charged with delivering stable accrual-based earnings will frequently be caught off guard by unexpected and financially crippling market price spikes and volume risk shocks. To defend effectively against these unpredictable events, that player must plan much further in advance and build a portfolio with the flexibility and robustness to survive the spot market environment.

The PAR approach provides the basis for analyzing future reporting periods-rather than being a snapshot of current financial risk, it requires the user to specify the period to be analyzed. (See Figure 2.) By planning well into the future, asset/retail-centric players can deliver consistent accrual-based earnings despite the challenges presented by volatile spot markets and volume risks.

Unlike VAR, PAR is not suited to a limit-based approach-if you wait for a limit to be breached, then you are probably too late to do much of anything about it. The most significant risks being managed, after all, are those due to major and unpredictable price and volume shocks and are not normally the result of trader actions. Also, most of the information about these events only arrives as delivery to the spot market approaches, and it is too late to do much more than react quickly-and hope that you are accurate-but with little time for detailed analysis and planning.

In practice, running a limit-based risk management approach for asset and retail portfolios results in limit alerts as risk management disasters begin to happen (though it is too late to do anything about them). To defend against these threats you must proactively build a robust portfolio well before the dangerous spot market arrives.

Setting it Up

With regard to organization structure, there is not yet an established standard for structuring MTM and accrual-focused risk management functions in the energy market or any other industry. There are conflicting arguments as to what constitutes a best practice in this arena.

One argument is that an organizations best skills and knowledge for doing any financial risk management resides in the wholesale trading operation, and therefore this is where all corporate or enterprise-wide risk management should occur. Another line of argument, occasionally used by dereg-ulating utilities, is that trading expertise resides in the corporate treasury function, so all wholesale trading and risk management should be done there.

Alternatively, many organizations conclude that the wholesale trading operation function should perform the MTM-based risk management of the trading portfolio, while a corporate function close to the chief financial officer should do the accrual-based management for the total corporate portfolio including asset and retail operations. This allows the trading operation to implement best practice business processes used by finance sector trading organizations, as if it were a bank with risk management centered on position, vAR, and credit risk. The finance department, with its expertise in accrual accounting (as used for corporate reporting), then can focus on PAR-based risk management for the enterprise.

Into Alignment

Utilities are taking action on a number of fronts to repair their battered credit ratings. Clearly, moves to conserve cashflow are crucial, but to restore faith in their business, utilities need to tailor their risk management methodology and accounting policy to reflect the underlying risk drivers of their business model. For those businesses that are exposed to volume risk and spot price volatility, this means predominantly using profit-at-risk measures and accrual accounting. Value-at-risk and mark-to-market accounting should be reserved for those parts of the business that are speculative.

By bringing about this alignment utilities will be able to provide ratings agencies with evidence of the transparency of their business, something that has, in some cases, been sorely lacking. Moreover, the increased insights utilities gain will allow them to maneuver efficiently through the risky energy jungle.

Greg Keers is chief strategy officer for KWI in London. PAR is trademarked and developed by KWI.

Copyright Edison Electric Institute Sep/Oct 2002

Provided by ProQuest Information and Learning Company. All rights Reserved