Email This Author Comment On Article About The Author More Articles By This Author

As the electricity market was being deregulated, the implicit assumption was that innovative solutions would emerge naturally as retailers used them for competitive differentiation and that a spectrum of price-risk-differentiated products and services would soon result. After all, competition should result naturally in cycles of innovation. We all certainly have seen these benefits in the deregulation of telephone service. Unfortunately, instead of provoking innovation, the combination of price to beat protections and boom-bust wholesale energy price volatility results in a game of electricity Old Maid. The company serving customers when these prices spike loses … and possibly even goes out of business. Rather than focus on the incorrect market designs and participant actions (which are deserving of criticism), this article examines what is really at work in the underlying economic systems. Economists rightly insist that demand will eventually bring about supply. Unfortunately, as the reader will see, the commonly used traditional corporate economic metrics can delay and possibly even prevent innovative moves because they fail miserably at accurately describing the situation at hand on two key points:

1. Flaw of averages: The common metrics use averages to describe things that are not fairly represented by the average. Volatility impacts must be characterized intrinsically by the metric if volatility impacts the decision (and we know it does). The old regulatory framework smoothed out the averages and hid them from market view. The intrinsic volatility risks were neither measured nor monetized.
2. Withstanding worst case scenarios: Conservatism tends to be compromised when people try to build a business case for a concept. Perhaps this is due to the nature of capital allocations in modern companies. Everyone wants their projects to be approved so everyone tries to sell their projects as better than the next person’s. Management seldom asks project sponsors to present their needs to withstand worst case scenarios for alternative business cases.

Why is a different treatment required? The perception of risk portrayal in traditional financial metrics was fair only when:

1. Risks were symmetric and averages reflected what could be expected, and
2. The likelihood and consequences of things going well was reasonably similar to the likelihood and consequences of things going poorly.

Unfortunately, today’s energy risks are asymmetric and better metrics are needed to properly alert management of consequent business risks. Compounding matters is the fact that the price risk in serving electricity customers was never “monetized” in the old regulatory paradigm and consequently missed the viewing screen of those attempting to open competitive electricity markets. In addition, the size of this risk proved to be large and highly uncertain … years of low risk followed by a season or two of high risks. Economists would suggest that risk pooling and price-risk differentiation of products and services should emerge. However, the reason they haven’t as of yet is all about The way risks like these are treated mathematically. Traditional Economic Metrics
How is it that well-meaning professionals can be so misled by seemingly solid economic predictions in today’s energy markets? The traditional metrics (net present value, return on investment, etc.) have been typically used as a basis for comparison among equally risky options rather than as an absolute index of their financial performance including these risk characteristics. Or, if there was a difference in the perception of risk among the options, one might be “discounted” by some percentage to make up for that risk premium. Risk itself was not characterized; it simply was used in some intuitive and emotional way to discount. One general rule in situations like this is to watch for key phrases in the dialogue. When you hear the phrase “I think” used in energy decision making you should consider that to be a warning sign that the person is speculating and taking on risk. On the other hand, when the phrase “what if” is asked, that person tends either to hedge or to dismiss the option as sufficiently unlikely that 1) it is justification to exit the business (which means the plan must include an “off ramp”) or 2) external protection is required (laying off the risk to others). While tiring to speculative types, this process of thinking through the risks is the stock and trade of the legal profession in writing up contracts between counterparties. Somehow we just need to formalize this into the economic evaluation itself. Said another way, the traditional metrics have been used for incremental decision making, not to decide whether a business is a good or bad idea in the first place. Part of the reason for this perspective is the implicit assumption that any ongoing decision in business is incremental, not disruptive. On the contrary, the impact of energy price volatility can be extremely disruptive. Valuing the Unlikely High Consequence Event
The reason people can be so cavalier and irresponsible is that the mathematics itself is so riddled with uncertainty that a business case becomes intrinsically argumentative. For example, consider what you would do if faced with the following mutually exclusive business propositions:

1. A 99.9% chance of earning $250, and a 0.1% chance of earning nothing
2. A 99% chance of earning $300, and a 1% chance of earning nothing
3. A 90% chance of earning $400, and a 10% chance of earning nothing
4. A 50/50 chance of earning $1,000, etc.
5. A 10% chance of earning $7,500, etc.
6. A 1% chance of earning $95,000, etc.
7. A 1 in 1,000 chance of earning $2,000,000, etc.

Exhibit 1 Expected Values

Which one would you recommend to your management? If there was going to be only one shot at this set of mutually exclusive opportunities, and a purely theoretical approach was offered, the statistician would suggest you simply multiply the odds times the payoff. As illustrated in Exhibit 1, the result would be the conclusion to take Option 7: a 1 in 1,000 chance at $2,000,000 since the expected value is $2,000 and that is so much higher than any of the other expected values. Hopefully we are not going to anger the reader by condemning such a silly notion, given there is only one shot at this game. The real pressing question for a business facing this situation today is whether you could withstand the management criticism to go even beyond options 2 or 3 in this list. At some point management will declare an event unlikely and will go with the other options, especially if the valuation and the odds both could be considered suspect or uncertain. This is where we see what is at the root of today’s failures to prepare for adverse economics when companies potentially face energy price spikes. It is simply too infrequent and too uncertain to clear the business-case criticisms, in part because they are using an expected value determination. The probabilities and the benefit valuation are viewed to be too subjective to give the company any confidence in the multiplication of the two effects, even where that clearly has a consequence on cumulative cash requirements in a business — surviving the worst case. Risk Management through Contingency Planning
The acid test of the combination of these factors is a conservative and realistically pessimistic view of business conditions on the cumulative cash-flow requirements. Can the company withstand that occurrence and does it have enough cash in the war chest to withstand that outcome? Few companies are this well capitalized, and the raft of bankruptcies in the past few months raises questions about whether many of today’s companies are adequately capitalized. How can that be, given all the really smart people in business today? Ironically, the root of all this is the raft of business school graduates who went to work on Wall Street and criticized these same companies for being too stodgy and conservative over the past few decades. Fed by comparisons to Enron, incorrect accounting reports, and often insidious relationships to the companies they were advising, the perversions seemed tolerable given that everyone appeared to be making money. While stockholders have every right to be angered by their loss of market valuation, the results shouldn’t have been all that surprising. Wall Street was rewarding companies that were leveraging their operations and avoiding asset-based strategies. Leverage increases rewards when times are good but also amplifies the negatives when times are bad. Unfortunately, the capital markets are attracted to rewards and are therefore unimpressed with energy infrastructure opportunities at the moment. So, we appear headed straight into a capital formation meltdown on the supply side of this business. Preventing: Heads I Win, Tails You Die

lign="right"> Exhibit 2: Electricity Price Cumulative Probability Distribution

The crass sounding comment in the subhead above is exactly how asymmetric risk works in energy markets today. Compare the two probability distributions shown here from a cumulative distribution perspective. Notice that the symmetry of the normal distribution and the right-hand tail of the log normal (which tends to be how all commodity prices behave). The result is that while both distributions have the same probability that prices will remain the same on average (shown at $50/MWh in this graph), the cumulative chance that the price will be $60 or lower is 10% under the normal distribution and about 30% under the log normal assumption. And, while the highest price might be tolerable in the normal distribution, it could be devastating in the log normal case. Said very simply, when things go wrong, they go very wrong, and it is not likely that the organization can “make it up in volume.” In fact, volume risk compounds the impact of price risk. The combined effects can be brutal. Then the real attribute worth noting should be the cumulative cash required to stay the course. We modeled this situation in a retail setting and compared the maximum negative cumulative cash flow requirements for the business in both hedged and unhedged positions. That is, how much cash (in thousands of dollars) was required to run this business for five years. As one might expect, the results are quite variable and can be illustrated best by using probability distributions for these cumulative cash flows, as shown in Exhibit 5. While the assumptions for this can be argued, notice that the general result is what you would expect intuitively. Hedging mitigates the worst case cash requirements; it does not affect the average as much. Said another way, hedging provides significant survivability protection. In Summary
Managers of today’s businesses that face this assymetric risk should seriously consider refining the ways they develop and evaluate their capital and operational plans. While averages may be intuitively attractive, and ranges may seem plausible, the capital allocation process invites assumption optimism. Business executives may not enjoy hearing about all the things that can go wrong, but if they persist in rewarding high-flying risk takers, they intrinsically expose their companies to consequent volatility and potential ruin. Slow and steady may not win the race, but “ready, fire, aim” is not a sure bet either. When options that have intrinsically different risk profiles are evaluated, we suggest evaluating them in the distributional sense to inform and instruct senior officers. Trying to hide the volatility does the organization serious disservice. And, it is important for the management team to recognize that doing nothing may be even riskier. It is probably instructive to construct the existing business case in the same model and look at how close the organization is to ruin if a few things fail to go as planned. Murphy is alive and well. Efforts to repeal the law have failed. And, yes, engineers like myself are criticized for being conservative and tending to point out things that can, and often do, go wrong. The reason we all do that is that we are trying to protect the organization. Temper the negatives if you will, but do not ignore them.